zlei6/ragflow
1
0
Fork 0
mirror of https://github.com/infiniflow/ragflow.git synced 2026-07-06 11:28:38 +08:00
Code Issues Packages Projects Releases Wiki Activity
Files
d770217b254fca92bcbc32fb5542e0ca88df100a
ragflow/uv.lock

10021 lines
826 KiB
TOML
Raw Normal View History

Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
version = 1
Go: refactor embedding interface (#14757) ### What problem does this PR solve? Provide embedding index according to the input text ### Type of change - [x] Refactoring --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2026-05-11 14:45:30 +08:00
revision = 3
feat(agent): Go port — canvas engine, 22 components, DSL v2, 13 endpoints (#15952) Ports the agent canvas subsystem from Python to Go. ## What's included ### Canvas Engine (Phase 0/1) - State engine, scheduler, variable resolver, Redis checkpoint store, cancel protocol - **209 tests** across canvas / component / io packages ### 22 Components (P0–P4) | Tier | Components | |---|---| | P0 T1+T2+T3 | LLM, Agent, ExitLoop, Switch, Categorize, Begin, Message, Invoke | | P1 T3 | VariableAggregator, VariableAssigner, StringTransform, ListOperations, DataOperations | | P2 T3 | Iteration, IterationItem, Loop, LoopItem | | P3 T3 | UserFillUp, Fillup | | P4 T5 | Browser, ExcelProcessor, DocsGenerator | ### DSL v2 Schema (Phase 2.5) - Typed v2 in-memory model with v1-to-v2 auto-detect converter - v1 legacy field stripping per plan §2.11.7 ### HTTP Endpoints & Bug Fixes (Plans PR1–PR3) - **DELETE SQL bug fix**: gorm v2 `Where("id = ?", id).Delete(...)` pattern - **CreateAgent validation**: title/DSL required, duplicate check, 103 envelope - **13 new endpoints**: templates, prompts, tags, sessions CRUD, chat/completions (SSE + non-stream stubs), rerun, test_db_connection, logs, webhook/logs - **756 Go unit tests** (745 → 756, +18) - **17 → 0 Python integration test failures** (test_agents.py + test_session_management/) ### Tools 21 eino tools: HTTPHelper, search tools, financial/data tools, mandatory stubs ### Infrastructure OTel observability, NATS message queue, DeepDoc gRPC client, SSRF guards, IDOR mitigation
2026-06-12 22:58:28 +08:00
requires-python = "==3.13.*"
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
resolution-markers = [
feat(agent): Go port — canvas engine, 22 components, DSL v2, 13 endpoints (#15952) Ports the agent canvas subsystem from Python to Go. ## What's included ### Canvas Engine (Phase 0/1) - State engine, scheduler, variable resolver, Redis checkpoint store, cancel protocol - **209 tests** across canvas / component / io packages ### 22 Components (P0–P4) | Tier | Components | |---|---| | P0 T1+T2+T3 | LLM, Agent, ExitLoop, Switch, Categorize, Begin, Message, Invoke | | P1 T3 | VariableAggregator, VariableAssigner, StringTransform, ListOperations, DataOperations | | P2 T3 | Iteration, IterationItem, Loop, LoopItem | | P3 T3 | UserFillUp, Fillup | | P4 T5 | Browser, ExcelProcessor, DocsGenerator | ### DSL v2 Schema (Phase 2.5) - Typed v2 in-memory model with v1-to-v2 auto-detect converter - v1 legacy field stripping per plan §2.11.7 ### HTTP Endpoints & Bug Fixes (Plans PR1–PR3) - **DELETE SQL bug fix**: gorm v2 `Where("id = ?", id).Delete(...)` pattern - **CreateAgent validation**: title/DSL required, duplicate check, 103 envelope - **13 new endpoints**: templates, prompts, tags, sessions CRUD, chat/completions (SSE + non-stream stubs), rerun, test_db_connection, logs, webhook/logs - **756 Go unit tests** (745 → 756, +18) - **17 → 0 Python integration test failures** (test_agents.py + test_session_management/) ### Tools 21 eino tools: HTTPHelper, search tools, financial/data tools, mandatory stubs ### Infrastructure OTel observability, NATS message queue, DeepDoc gRPC client, SSRF guards, IDOR mitigation
2026-06-12 22:58:28 +08:00
"sys_platform == 'darwin'",
"platform_machine == 'aarch64' and sys_platform == 'linux'",
"(platform_machine != 'aarch64' and sys_platform == 'linux') or (sys_platform != 'darwin' and sys_platform != 'linux')",
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Fix: upgrade pyasn1 from 0.6.2 to 0.6.3 to address CVE-2026-30922 (#13773) ## Summary - Adds `pyasn1>=0.6.3` as a `[tool.uv.constraint-dependencies]` entry to mitigate **CVE-2026-30922** (CVSS 7.5 HIGH) - Regenerates `uv.lock` so the resolved pyasn1 version moves from **0.6.2 to 0.6.3** ## Details **CVE-2026-30922** is a Denial of Service vulnerability in pyasn1 caused by unbounded recursion when decoding ASN.1 data with deeply nested structures. An attacker can send crafted payloads with thousands of nested SEQUENCE or SET tags to trigger a `RecursionError` crash or memory exhaustion. - **Severity:** HIGH (CVSS 7.5) - **Affected versions:** pyasn1 < 0.6.3 - **Fixed in:** pyasn1 >= 0.6.3 - **NVD:** https://nvd.nist.gov/vuln/detail/CVE-2026-25769 `pyasn1` is not a direct dependency of RAGFlow but is pulled in transitively via `google-auth` -> `rsa` -> `pyasn1-modules` -> `pyasn1`. The `constraint-dependencies` mechanism in uv is the correct way to enforce a minimum version for transitive dependencies without polluting the direct dependency list. ## Test plan - [x] `pyproject.toml` passes TOML validation - [x] `uv lock` resolves successfully with the new constraint - [x] pyasn1 version in `uv.lock` is now 0.6.3 - [ ] Existing CI/CD tests continue to pass Closes #13686
2026-03-26 19:37:34 -07:00
[manifest]
feat: bump Python minimum from 3.12 to 3.13, drop strenum backport (#14767) Closes #14753 ## What changed | File | Change | |---|---| | `pyproject.toml` | `requires-python` → `>=3.13,<3.15`; remove `strenum==0.4.15` | | `Dockerfile` | `uv python install 3.13`, `uv sync --python 3.13` | | `.github/workflows/tests.yml` | `uv sync --python 3.13` on both matrix legs | | `CLAUDE.md` | dev setup command + requirements note updated | | `deepdoc/parser/mineru_parser.py` | `from strenum import StrEnum` → `from enum import StrEnum` | | `agent/tools/code_exec.py` | same | `StrEnum` has been in the stdlib since Python 3.11 — the `strenum` backport package is no longer needed once the floor is 3.13. ## Why uv.lock is not regenerated `uv lock --python 3.13` fails because: 1. The infiniflow/graspologic fork pins `numpy>=1.26.4,<2.0.0` 2. `tensorflow-cpu>=2.20.0` (the first release with cp313 wheels) depends on `ml-dtypes>=0.5.1`, which requires `numpy>=2.1.0` 3. These two constraints are irreconcilable on Python 3.13 The lockfile regeneration requires loosening the `numpy` upper bound in the `infiniflow/graspologic` fork. Once that fork commit is updated and the SHA in `pyproject.toml:49` is bumped, `uv lock --python 3.13` will succeed. ## RFC corrections Two claims in the original RFC (#14753) did not hold up under code review: - **"graspologic hard-blocks 3.13"** — the infiniflow fork at the pinned commit has no `<3.13` Python constraint. The blocker is the transitive `numpy<2.0.0` conflict with tensorflow-cpu's test dependency, not a direct Python version cap. - **"free-threading throughput gains for I/O-bound workload"** — Python 3.13 free-threading requires a special `--disable-gil` build and provides no benefit for async I/O code (the GIL is already released during I/O). The real motivation is forward compatibility and improved error messages.
2026-05-15 08:40:53 +02:00
constraints = [
{ name = "pyasn1", specifier = ">=0.6.3" },
{ name = "trio", specifier = ">=0.26.0", index = "https://pypi.org/simple" },
Fix: #6098 - Add validation logic for parser_config when update document (#13911) ### What problem does this PR solve? Add validation logic for parser_config. Refactor the processing flow. Before change, validation logics and update logics are mixed up - some validation logis executes followed by some update logic executes and then another such "validation-and-then-update" which is not good. After change, all validation logic executes firstly. Update logic will be executed after ALL validation logic executed. Validation logic for parameters (that come from front end) will be checked using Pydantic. For validation logic that depends on data from DB, they will be in separate methods. ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue) - [x] Refactoring
2026-04-07 11:33:05 +08:00
]
feat: bump Python minimum from 3.12 to 3.13, drop strenum backport (#14767) Closes #14753 ## What changed | File | Change | |---|---| | `pyproject.toml` | `requires-python` → `>=3.13,<3.15`; remove `strenum==0.4.15` | | `Dockerfile` | `uv python install 3.13`, `uv sync --python 3.13` | | `.github/workflows/tests.yml` | `uv sync --python 3.13` on both matrix legs | | `CLAUDE.md` | dev setup command + requirements note updated | | `deepdoc/parser/mineru_parser.py` | `from strenum import StrEnum` → `from enum import StrEnum` | | `agent/tools/code_exec.py` | same | `StrEnum` has been in the stdlib since Python 3.11 — the `strenum` backport package is no longer needed once the floor is 3.13. ## Why uv.lock is not regenerated `uv lock --python 3.13` fails because: 1. The infiniflow/graspologic fork pins `numpy>=1.26.4,<2.0.0` 2. `tensorflow-cpu>=2.20.0` (the first release with cp313 wheels) depends on `ml-dtypes>=0.5.1`, which requires `numpy>=2.1.0` 3. These two constraints are irreconcilable on Python 3.13 The lockfile regeneration requires loosening the `numpy` upper bound in the `infiniflow/graspologic` fork. Once that fork commit is updated and the SHA in `pyproject.toml:49` is bumped, `uv lock --python 3.13` will succeed. ## RFC corrections Two claims in the original RFC (#14753) did not hold up under code review: - **"graspologic hard-blocks 3.13"** — the infiniflow fork at the pinned commit has no `<3.13` Python constraint. The blocker is the transitive `numpy<2.0.0` conflict with tensorflow-cpu's test dependency, not a direct Python version cap. - **"free-threading throughput gains for I/O-bound workload"** — Python 3.13 free-threading requires a special `--disable-gil` build and provides no benefit for async I/O code (the GIL is already released during I/O). The real motivation is forward compatibility and improved error messages.
2026-05-15 08:40:53 +02:00
overrides = [{ name = "attrs", specifier = ">=23.2.0" }]
build(deps): bump crawl4ai from 0.8.9 to 0.9.0 (#16470)
2026-06-30 09:34:48 +08:00
excludes = [
"agentrun-mem0ai",
"unclecode-litellm",
]
Fix: #6098 - Add validation logic for parser_config when update document (#13911) ### What problem does this PR solve? Add validation logic for parser_config. Refactor the processing flow. Before change, validation logics and update logics are mixed up - some validation logis executes followed by some update logic executes and then another such "validation-and-then-update" which is not good. After change, all validation logic executes firstly. Update logic will be executed after ALL validation logic executed. Validation logic for parameters (that come from front end) will be checked using Pydantic. For validation logic that depends on data from DB, they will be in separate methods. ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue) - [x] Refactoring
2026-04-07 11:33:05 +08:00
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
[[package]]
name = "aenum"
version = "3.1.17"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/07/e9/8b283567c1fef7c24d1f390b37daede8b61593d8cdaffb8e95d571699e83/aenum-3.1.17.tar.gz", hash = "sha256:a969a4516b194895de72c875ece355f17c0d272146f7fda346ef74f93cf4d5ba" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/48/8d/1fe30c6fd8999b9d462547c4a1bb6690bda24af38f2913c4bec7decb81f2/aenum-3.1.17-py3-none-any.whl", hash = "sha256:8b883a37a04e74cc838ac442bdd28c266eae5bbf13e1342c7ef123ed25230139" },
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
]
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
[[package]]
name = "agentrun-sdk"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
version = "0.0.51"
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
dependencies = [
{ name = "alibabacloud-agentrun20250910" },
{ name = "alibabacloud-bailian20231229" },
{ name = "alibabacloud-devs20230714" },
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ name = "alibabacloud-gpdb20160503" },
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
{ name = "alibabacloud-tea-openapi" },
{ name = "crcmod" },
{ name = "httpx" },
{ name = "litellm" },
{ name = "pydantic" },
{ name = "pydash" },
{ name = "python-dotenv" },
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ name = "tablestore-agent-storage" },
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
{ name = "typing-extensions" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/55/56/7caca0780cee6098c39f2739ed9867746afb01e22d5a8bf22d353d8920b2/agentrun_sdk-0.0.51.tar.gz", hash = "sha256:48ec49fd925bb204d1565488f832fc0a53cd8304938274f0ab528d5a57cdb1d6" }
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/78/fe/a06e34120e03d09f38485098b9d2704d4d217d7acfdeedd1731bcd8856a9/agentrun_sdk-0.0.51-py3-none-any.whl", hash = "sha256:b33c4b5ff7cb1a2987dca34e53b2fa959f103ea0e6256c83017a0f82b744382e" },
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
]
Refa: migrate chat models to LiteLLM (#9394) ### What problem does this PR solve? All models pass the mock response tests, which means that if a model can return the correct response, everything should work as expected. However, not all models have been fully tested in a real environment, the real API_KEY. I suggest actively monitoring the refactored models over the coming period to ensure they work correctly and fixing them step by step, or waiting to merge until most have been tested in practical environment. ### Type of change - [x] Refactoring
2025-08-12 10:59:20 +08:00
[[package]]
name = "aiofiles"
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
version = "24.1.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/0b/03/a88171e277e8caa88a4c77808c20ebb04ba74cc4681bf1e9416c862de237/aiofiles-24.1.0.tar.gz", hash = "sha256:22a075c9e5a3810f0c2e48f3008c94d68c65d763b9b03857924c99e57355166c" }
Refa: migrate chat models to LiteLLM (#9394) ### What problem does this PR solve? All models pass the mock response tests, which means that if a model can return the correct response, everything should work as expected. However, not all models have been fully tested in a real environment, the real API_KEY. I suggest actively monitoring the refactored models over the coming period to ensure they work correctly and fixing them step by step, or waiting to merge until most have been tested in practical environment. ### Type of change - [x] Refactoring
2025-08-12 10:59:20 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/a5/45/30bb92d442636f570cb5651bc661f52b610e2eec3f891a5dc3a4c3667db0/aiofiles-24.1.0-py3-none-any.whl", hash = "sha256:b4ec55f4195e3eb5d7abd1bf7e061763e864dd4954231fb8539a0ef8bb8260e5" },
Refa: migrate chat models to LiteLLM (#9394) ### What problem does this PR solve? All models pass the mock response tests, which means that if a model can return the correct response, everything should work as expected. However, not all models have been fully tested in a real environment, the real API_KEY. I suggest actively monitoring the refactored models over the coming period to ensure they work correctly and fixing them step by step, or waiting to merge until most have been tested in practical environment. ### Type of change - [x] Refactoring
2025-08-12 10:59:20 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "aiohappyeyeballs"
DOCS: for release. (#6023) ### What problem does this PR solve? ### Type of change - [x] Documentation Update
2025-03-13 15:09:29 +08:00
version = "2.6.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/26/30/f84a107a9c4331c14b2b586036f40965c128aa4fee4dda5d3d51cb14ad54/aiohappyeyeballs-2.6.1.tar.gz", hash = "sha256:c3f9d0113123803ccadfdf3f0faa505bc78e6a72d1cc4806cbd719826e943558" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/0f/15/5bf3b99495fb160b63f95972b81750f18f7f4e02ad051373b669d17d44f2/aiohappyeyeballs-2.6.1-py3-none-any.whl", hash = "sha256:f349ba8f4b75cb25c99c5c2d84e997e485204d2902a9597802b0371f09331fb8" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "aiohttp"
build(deps): bump aiohttp from 3.13.3 to 3.14.1 (#16090)
2026-06-16 20:07:32 +08:00
version = "3.14.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "aiohappyeyeballs" },
{ name = "aiosignal" },
{ name = "attrs" },
{ name = "frozenlist" },
{ name = "multidict" },
{ name = "propcache" },
{ name = "yarl" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/82/78/8ea7308cac6934de8c74a14f3d5f65d1c89287426688be79538d0e5c013d/aiohttp-3.14.1.tar.gz", hash = "sha256:307f2cff90a764d329e77040603fa032db89c5c24fdad50c4c15334cba744035" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/bc/97/bd137012dd97e1649162b099135a80e1fd59aaa807b2430fc448d1029aff/aiohttp-3.14.1-cp313-cp313-android_21_arm64_v8a.whl", hash = "sha256:b3a03285a7f9c7b016324574a6d92a1c895da6b978cb8f1deee3ac72bc6da178" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ef/79/e5cc690e9d922a66887ceeaca53a8ffd5a7b0be3816142b7abc433742d89/aiohttp-3.14.1-cp313-cp313-android_21_x86_64.whl", hash = "sha256:2a73f487ab8ef5abbb24b7aa9b73e98eaba9e9e031804ff2416f02eca315ccaf" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fe/22/a73ccbf9dbd6e26dda0b24d5fd5db7da92ee3383a79f47677ffb834c5c5b/aiohttp-3.14.1-cp313-cp313-ios_13_0_arm64_iphoneos.whl", hash = "sha256:915fbb7b41b115192259f8c9ae58f3ddc444d2b5579917270211858e606a4afd" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3b/b9/57ed8eaf596321c2ad747bd480fb1700dbd7177c60dfc9e4c187f629662e/aiohttp-3.14.1-cp313-cp313-ios_13_0_arm64_iphonesimulator.whl", hash = "sha256:7fb4bdf95b0561a79f259f9d28fbc109728c5ee7f27aff6391f0ca703a329abe" },
{ url = "https://mirrors.aliyun.com/pypi/packages/78/c0/5ebe5270a7c140d7c6f79dcb018640225f14d406c149e4eec04a7d82fe71/aiohttp-3.14.1-cp313-cp313-ios_13_0_x86_64_iphonesimulator.whl", hash = "sha256:1b9748363260121d2927704f5d4fc498150669ca3ae93625986ee89c8f80dcd4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/75/7f/8cdaa24fc7983865e0915153b96a9ac5bcdd3548d64c5a27d17cecccad2d/aiohttp-3.14.1-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:86a6dab78b0e43e2897a3bbe15745aa60dc5423ca437b7b0b164c069bf91b876" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b2/f4/c4227aacfacc5cb0cc2d119b65301d177912a6842cd64e120c47af76064f/aiohttp-3.14.1-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:4dfd6e47d3c44c2279907607f73a4240b88c69eb8b90da7e2441a8045dfd21da" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ab/01/a2d5f96cd4e74424864d30bc0a7e44d0a12dacdcfa91b5b2d1bd3dca6bf3/aiohttp-3.14.1-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:317acd9f8602858dc7d59679812c376c7f0b97bcbbf16e0d6237f54141d8a8a6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e8/ed/3c0fb5c500fdd8e7ebc10d1889c04384fffa1a9163eac1356088ca9da1b1/aiohttp-3.14.1-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:bd869c427324e5cb15195793de951295710db28be7d818247f3097b4ab5d4b96" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0b/ab/d4c924d9bd5be3050c226612413ce68cb54c70d2c31b661bfc8d9a5b6a70/aiohttp-3.14.1-cp313-cp313-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:93b032b5ec3255473c143627d21a69ac74ae12f7f33974cb587c564d11b1066f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/19/2a/37326821ff779084020cdc33224d20b19f42f4183a500ff92022a739eda7/aiohttp-3.14.1-cp313-cp313-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:f234b4deb12f3ad59127e037bc57c40c21e45b45282df7d3a55a0f409f595296" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b3/4f/6e947ba73e4ce09070761c05ed3a8ceb7c21f5e46798671d8b2aac0e4626/aiohttp-3.14.1-cp313-cp313-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:9af6779bfb46abf124068327abcdf9ce95c9ef8287a3e8da76ccf2d0f16c28fa" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9d/6e/dbf1d0625dc711fb2851f4f3c3055c39ed58bae92082d8c627dbe6013736/aiohttp-3.14.1-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:faccab372e66bc76d5731525e7f1143c922271725b9d38c9f97edcc66266b451" },
{ url = "https://mirrors.aliyun.com/pypi/packages/44/c2/5e25098a67268ed369483ae7d1a58bd0a13d03aab860d2a0e4a6eb25b046/aiohttp-3.14.1-cp313-cp313-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:f380468b09d2a81633ee863b0ec5648d364bd17bb8ecfb8c2f387f7ac1faf42c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2a/bd/cf9cee17e140f942a3de73e658a543aa8fbf35a5fc67a9d2538d52d77f0b/aiohttp-3.14.1-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:97e704dcd26271f5bda3fa07c3ce0fb76d6d3f8659f4baa1a24442cc9ba177ca" },
{ url = "https://mirrors.aliyun.com/pypi/packages/89/6d/5684f8c59045c96f81a18cefbc1fbbd79d25b88f1c622f2a5c5c08fcb632/aiohttp-3.14.1-cp313-cp313-musllinux_1_2_armv7l.whl", hash = "sha256:269b76ac5394092b95bc4a098f4fc6c191c083c3bd12775d1e30e663132f6a09" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a8/40/35caf3170f8359760740a7d9aa0fff2e344bef98e1d1186f5a0f6dec17e6/aiohttp-3.14.1-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:5c0b3e614340c889d575451696374c9d17affd54cd607ca0babed8f8c37b9397" },
{ url = "https://mirrors.aliyun.com/pypi/packages/6d/a1/b0c61e7a137f0d81de49a82023a6df73c3c16d6fefb0f8e4a93d21639002/aiohttp-3.14.1-cp313-cp313-musllinux_1_2_riscv64.whl", hash = "sha256:5663ee9257cfa1add7253a7da3035a02f31b6600ec48261585e1800a81533080" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0b/41/194ea4623693009fcefebef7aef63c141754f153e9cd0d39d3b9e36c175c/aiohttp-3.14.1-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:603a2c834142172ffddc054067f5ec0ca65d57a0aa98a71bc81952573208e345" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ba/45/4de841f005cfe1fd63e2a2fe011262c515e2a62aa6994b15947e7d717ac9/aiohttp-3.14.1-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:cb21957bb8aca671c1765e32f58164cf0c50e6bf41c0bbbd16da20732ecaf588" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e4/ae/dbce10533d3896d544d5053939ed75b7dc31a1b0973d959b1b5ae21028d6/aiohttp-3.14.1-cp313-cp313-win32.whl", hash = "sha256:e509a55f681e6158c20f70f102f9cf61fb20fbc382272bc6d94b7343f2582780" },
{ url = "https://mirrors.aliyun.com/pypi/packages/7b/d9/0bf1a19362c32f06229da5e7ddfcec91f93474d6307f7a2d3135e9c674dc/aiohttp-3.14.1-cp313-cp313-win_amd64.whl", hash = "sha256:1ac8531b638959718e18c2207fbfe297819875da46a740b29dfa29beba64355a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/22/0a/62e7232dc9484fbec112ceb32efb6a624cc7994ec6e2b019286f17c4e8f2/aiohttp-3.14.1-cp313-cp313-win_arm64.whl", hash = "sha256:250d14af67f6b6a1a4a811049b1afa69d61d617fca6bf33149b3ab1a6dbcf7b8" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "aiolimiter"
version = "1.2.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f1/23/b52debf471f7a1e42e362d959a3982bdcb4fe13a5d46e63d28868807a79c/aiolimiter-1.2.1.tar.gz", hash = "sha256:e02a37ea1a855d9e832252a105420ad4d15011505512a1a1d814647451b5cca9" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/f3/ba/df6e8e1045aebc4778d19b8a3a9bc1808adb1619ba94ca354d9ba17d86c3/aiolimiter-1.2.1-py3-none-any.whl", hash = "sha256:d3f249e9059a20badcb56b61601a83556133655c11d1eb3dd3e04ff069e5f3c7" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
feat: add OceanBase doc engine (#11228) ### What problem does this PR solve? Add OceanBase doc engine. Close #5350 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-20 10:00:14 +08:00
[[package]]
name = "aiomysql"
version = "0.3.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
feat: add OceanBase doc engine (#11228) ### What problem does this PR solve? Add OceanBase doc engine. Close #5350 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-20 10:00:14 +08:00
dependencies = [
{ name = "pymysql" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/29/e0/302aeffe8d90853556f47f3106b89c16cc2ec2a4d269bdfd82e3f4ae12cc/aiomysql-0.3.2.tar.gz", hash = "sha256:72d15ef5cfc34c03468eb41e1b90adb9fd9347b0b589114bd23ead569a02ac1a" }
feat: add OceanBase doc engine (#11228) ### What problem does this PR solve? Add OceanBase doc engine. Close #5350 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-20 10:00:14 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/4c/af/aae0153c3e28712adaf462328f6c7a3c196a1c1c27b491de4377dd3e6b52/aiomysql-0.3.2-py3-none-any.whl", hash = "sha256:c82c5ba04137d7afd5c693a258bea8ead2aad77101668044143a991e04632eb2" },
feat: add OceanBase doc engine (#11228) ### What problem does this PR solve? Add OceanBase doc engine. Close #5350 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-20 10:00:14 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "aiosignal"
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
version = "1.4.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "frozenlist" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/61/62/06741b579156360248d1ec624842ad0edf697050bbaf7c3e46394e106ad1/aiosignal-1.4.0.tar.gz", hash = "sha256:f47eecd9468083c2029cc99945502cb7708b082c232f9aca65da147157b251c7" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/fb/76/641ae371508676492379f16e2fa48f4e2c11741bd63c48be4b12a6b09cba/aiosignal-1.4.0-py3-none-any.whl", hash = "sha256:053243f8b92b990551949e63930a839ff0cf0b0ebbe0597b0f3fb19e1a0fe82e" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: Add box connector (#11845) ### What problem does this PR solve? Feat: Add box connector ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-12 10:23:40 +08:00
[[package]]
name = "aiosmtplib"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "5.0.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a2/15/c2dc93a58d716bce64b53918d3cf667d86c96a56a9f3a239a9f104643637/aiosmtplib-5.0.0.tar.gz", hash = "sha256:514ac11c31cb767c764077eb3c2eb2ae48df6f63f1e847aeb36119c4fc42b52d" }
Feat: Add box connector (#11845) ### What problem does this PR solve? Feat: Add box connector ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-12 10:23:40 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/99/42/b997c306dc54e6ac62a251787f6b5ec730797eea08e0336d8f0d7b899d5f/aiosmtplib-5.0.0-py3-none-any.whl", hash = "sha256:95eb0f81189780845363ab0627e7f130bca2d0060d46cd3eeb459f066eb7df32" },
Feat: Add box connector (#11845) ### What problem does this PR solve? Feat: Add box connector ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-12 10:23:40 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "aiosqlite"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "0.22.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/4e/8a/64761f4005f17809769d23e518d915db74e6310474e733e3593cfc854ef1/aiosqlite-0.22.1.tar.gz", hash = "sha256:043e0bd78d32888c0a9ca90fc788b38796843360c855a7262a532813133a0650" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/00/b7/e3bf5133d697a08128598c8d0abc5e16377b51465a33756de24fa7dee953/aiosqlite-0.22.1-py3-none-any.whl", hash = "sha256:21c002eb13823fad740196c5a2e9d8e62f6243bd9e7e4a1f87fb5e44ecb4fceb" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "akracer"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "0.0.14"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/1e/c6/f38feed5b961d73e1b4cb049fdb45338356e0f5b828b230c00d0e51f3137/akracer-0.0.14.tar.gz", hash = "sha256:e084c14bf6d9a02d5da375e3af1cba3d46f103aa1cf3a2010593b3e95bf1c29a" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/53/cb/1041355b14cd4b76ac082e8c676858f6eddb78f0ba37c59284adf36e5103/akracer-0.0.14-py3-none-any.whl", hash = "sha256:629eaccd0e1d18366804b797eb2692ed47bed0028f55b5a5af3cc277d521df04" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "akshare"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.18.10"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
{ name = "aiohttp" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "akracer", marker = "sys_platform == 'linux'" },
{ name = "beautifulsoup4" },
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
{ name = "curl-cffi" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "decorator" },
{ name = "html5lib" },
{ name = "jsonpath" },
{ name = "lxml" },
{ name = "mini-racer", marker = "sys_platform != 'linux'" },
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
{ name = "nest-asyncio" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "openpyxl" },
{ name = "pandas" },
{ name = "py-mini-racer", marker = "sys_platform == 'linux'" },
{ name = "requests" },
{ name = "tabulate" },
{ name = "tqdm" },
{ name = "urllib3" },
{ name = "xlrd" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/fa/04/9a43970c8e19c28d697681ad79139e04e1f42e89b21cc5b9e20a84e3f2f7/akshare-1.18.10.tar.gz", hash = "sha256:992554fafc5a4099bc005189422850d6d27042f83c197056168514ce1b1ecdf4" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/d9/55/9615bd8b8c51df8ea833291b96e848eeaac7e08273503fe94ac56c4b4754/akshare-1.18.10-py3-none-any.whl", hash = "sha256:258ab5f97309bc70f017ca070a65338f9473e5df961c15ba300966eef93702cb" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "alabaster"
version = "1.0.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a6/f8/d9c74d0daf3f742840fd818d69cfae176fa332022fd44e3469487d5a9420/alabaster-1.0.0.tar.gz", hash = "sha256:c00dca57bca26fa62a6d7d0a9fcce65f3e026e9bfe33e9c538fd3fbb2144fd9e" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/7e/b3/6b4067be973ae96ba0d615946e314c5ae35f9f993eca561b356540bb0c2b/alabaster-1.0.0-py3-none-any.whl", hash = "sha256:fc6786402dc3fcb2de3cabd5fe455a2db534b371124f1f21de8731783dec828b" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
[[package]]
name = "alibabacloud-agentrun20250910"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
version = "5.7.4"
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
dependencies = [
{ name = "alibabacloud-tea-openapi" },
{ name = "darabonba-core" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/c3/27/c50eb2ecfea19f0703581c278e65dd294467b198c08a330032e92734303a/alibabacloud_agentrun20250910-5.7.4.tar.gz", hash = "sha256:3e4d83619a7b7573ca3967fb6cac34e8442ecfceb15607a31bea48e217941932" }
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/8a/b3/28d571a6fb6ea710f3769ad9b7e0e1e74dc6e6abacb9e0fc72b5ffed8bd6/alibabacloud_agentrun20250910-5.7.4-py3-none-any.whl", hash = "sha256:fbfcb01cde85ce9c655c62e87fab75d33047bc391c9598b6d42fb1d773c5c3cf" },
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
]
[[package]]
name = "alibabacloud-bailian20231229"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "2.8.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
dependencies = [
{ name = "alibabacloud-tea-openapi" },
{ name = "darabonba-core" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/b4/65/2aee1e58bb3eec52c4892637ee15c453b0a3c7797b9b68f49bb5e9dd4e60/alibabacloud_bailian20231229-2.8.1.tar.gz", hash = "sha256:d39a79cc11b7bd0cd59054b0c8a943923f4f3330da243c83446524aab4b63ed8" }
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/50/c6/97c771aa4305844c20549c61f79dc48ed418838fed77608240475f4d53cb/alibabacloud_bailian20231229-2.8.1-py3-none-any.whl", hash = "sha256:403678010e65412ee5f0f80c2a831bb50d5e4178f9e616c21fc2793232f25913" },
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
]
[[package]]
name = "alibabacloud-credentials"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.0.7"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
dependencies = [
{ name = "aiofiles" },
{ name = "alibabacloud-credentials-api" },
{ name = "alibabacloud-tea" },
{ name = "apscheduler" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/3c/2b/596a8b2cb6d08a75a6c85a98996d2a6f3a43a40aea5f892728bfce025b54/alibabacloud_credentials-1.0.7.tar.gz", hash = "sha256:80428280b4bcf95461d41d1490a22360b8b67d1829bf1eb38f74fabcc693f1b3" }
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/25/86/f8dbcc689d6f4ba0e1e709a9b401b633052138daf20f7ce661c073a45823/alibabacloud_credentials-1.0.7-py3-none-any.whl", hash = "sha256:465c779cfa284e8900c08880d764197289b1edd4c72c0087c3effe6bb2b4dea3" },
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
]
[[package]]
name = "alibabacloud-credentials-api"
version = "1.0.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a0/87/1d7019d23891897cb076b2f7e3c81ab3c2ba91de3bb067196f675d60d34c/alibabacloud-credentials-api-1.0.0.tar.gz", hash = "sha256:8c340038d904f0218d7214a8f4088c31912bfcf279af2cbc7d9be4897a97dd2f" }
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
[[package]]
name = "alibabacloud-devs20230714"
version = "2.4.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
dependencies = [
{ name = "alibabacloud-endpoint-util" },
{ name = "alibabacloud-openapi-util" },
{ name = "alibabacloud-tea-openapi" },
{ name = "alibabacloud-tea-util" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a0/b4/a6425a5d54dbdd83206b9c0418e9fded4764a1125bbefbe9ff9511ed2a72/alibabacloud_devs20230714-2.4.1.tar.gz", hash = "sha256:461e7614dc382b49d576ac8713d949beb48b1979cea002922bdb284883360f20" }
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/1b/c6/7d375cc1b1cab0f46950f556b70a2b17235747429a0889b73f3d46ff6023/alibabacloud_devs20230714-2.4.1-py3-none-any.whl", hash = "sha256:dbd260718e6db50021d804218b40bc99ee9c7e40b1def382aef8e542f5921113" },
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
]
Feat: add DingTalk AI Table connector and integration for data synch… (#13413) ### What problem does this PR solve? Add DingTalk AI Table connector and integration for data synchronization Issue #13400 ### Type of change - [x] New Feature (non-breaking change which adds functionality) Co-authored-by: wangheyang <wangheyang@corp.netease.com>
2026-03-06 21:13:23 +08:00
[[package]]
name = "alibabacloud-dingtalk"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "2.2.48"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: add DingTalk AI Table connector and integration for data synch… (#13413) ### What problem does this PR solve? Add DingTalk AI Table connector and integration for data synchronization Issue #13400 ### Type of change - [x] New Feature (non-breaking change which adds functionality) Co-authored-by: wangheyang <wangheyang@corp.netease.com>
2026-03-06 21:13:23 +08:00
dependencies = [
{ name = "alibabacloud-endpoint-util" },
{ name = "alibabacloud-gateway-dingtalk" },
{ name = "alibabacloud-gateway-spi" },
{ name = "alibabacloud-openapi-util" },
{ name = "alibabacloud-tea-openapi" },
{ name = "alibabacloud-tea-util" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/77/bd/99b3a8ec49de74d845943d8525eb55b7262b749a87025177d4fe1cad0e1f/alibabacloud_dingtalk-2.2.48.tar.gz", hash = "sha256:99cfc1b4d320750f8c2eb8fdea4b58e2737155de2ffee9313279cd0e28108722" }
Feat: add DingTalk AI Table connector and integration for data synch… (#13413) ### What problem does this PR solve? Add DingTalk AI Table connector and integration for data synchronization Issue #13400 ### Type of change - [x] New Feature (non-breaking change which adds functionality) Co-authored-by: wangheyang <wangheyang@corp.netease.com>
2026-03-06 21:13:23 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/1a/28/1939243ec148dc59172c2badd3c7a9f55fb7652bd8b22add349ca07dffa8/alibabacloud_dingtalk-2.2.48-py3-none-any.whl", hash = "sha256:8b953320b919a08463d645d24d02698a292df0f44b88d7d6ae26db528d73c882" },
Feat: add DingTalk AI Table connector and integration for data synch… (#13413) ### What problem does this PR solve? Add DingTalk AI Table connector and integration for data synchronization Issue #13400 ### Type of change - [x] New Feature (non-breaking change which adds functionality) Co-authored-by: wangheyang <wangheyang@corp.netease.com>
2026-03-06 21:13:23 +08:00
]
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
[[package]]
name = "alibabacloud-endpoint-util"
version = "0.0.4"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/92/7d/8cc92a95c920e344835b005af6ea45a0db98763ad6ad19299d26892e6c8d/alibabacloud_endpoint_util-0.0.4.tar.gz", hash = "sha256:a593eb8ddd8168d5dc2216cd33111b144f9189fcd6e9ca20e48f358a739bbf90" }
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
Feat: add DingTalk AI Table connector and integration for data synch… (#13413) ### What problem does this PR solve? Add DingTalk AI Table connector and integration for data synchronization Issue #13400 ### Type of change - [x] New Feature (non-breaking change which adds functionality) Co-authored-by: wangheyang <wangheyang@corp.netease.com>
2026-03-06 21:13:23 +08:00
[[package]]
name = "alibabacloud-gateway-dingtalk"
version = "1.0.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: add DingTalk AI Table connector and integration for data synch… (#13413) ### What problem does this PR solve? Add DingTalk AI Table connector and integration for data synchronization Issue #13400 ### Type of change - [x] New Feature (non-breaking change which adds functionality) Co-authored-by: wangheyang <wangheyang@corp.netease.com>
2026-03-06 21:13:23 +08:00
dependencies = [
{ name = "alibabacloud-gateway-spi" },
{ name = "alibabacloud-tea-util" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/d2/40/751d8bdf133d7fcf053f10c98e8e506810e7bee06458a02eaaa14d30ac26/alibabacloud_gateway_dingtalk-1.0.2.tar.gz", hash = "sha256:acea8b0b1d11e0394913f0b0899ddd19c0bfceab716060449b57fcc250ceb300" }
Feat: add DingTalk AI Table connector and integration for data synch… (#13413) ### What problem does this PR solve? Add DingTalk AI Table connector and integration for data synchronization Issue #13400 ### Type of change - [x] New Feature (non-breaking change which adds functionality) Co-authored-by: wangheyang <wangheyang@corp.netease.com>
2026-03-06 21:13:23 +08:00
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
[[package]]
name = "alibabacloud-gateway-spi"
version = "0.0.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
dependencies = [
{ name = "alibabacloud-credentials" },
]
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ab/98/d7111245f17935bf72ee9bea60bbbeff2bc42cdfe24d2544db52bc517e1a/alibabacloud_gateway_spi-0.0.3.tar.gz", hash = "sha256:10d1c53a3fc5f87915fbd6b4985b98338a776e9b44a0263f56643c5048223b8b" }
[[package]]
name = "alibabacloud-gpdb20160503"
version = "5.5.0"
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
dependencies = [
{ name = "alibabacloud-credentials" },
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ name = "alibabacloud-tea-openapi" },
{ name = "darabonba-core" },
]
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/73/9a/0393a3294fae7b15f12f5b03e61676c5301d2ef0aaddf6c03c02e9610447/alibabacloud_gpdb20160503-5.5.0.tar.gz", hash = "sha256:f7060206863066a63bbe70a7cfa32fe6c08dc4dd82f3547941e2ce528488dc0f" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/d8/5c/771957c85dbb1949ff48d4ba425aa51441260a175fa485c0695230b1ae4a/alibabacloud_gpdb20160503-5.5.0-py3-none-any.whl", hash = "sha256:0e1995d0fa32d71f201e899fa73c5e250382826c70676283209ca6ea0f0017c2" },
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
]
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
[[package]]
name = "alibabacloud-openapi-util"
version = "0.2.4"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
dependencies = [
{ name = "alibabacloud-tea-util" },
{ name = "cryptography" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f6/51/be5802851a4ed20ac2c6db50ac8354a6e431e93db6e714ca39b50983626f/alibabacloud_openapi_util-0.2.4.tar.gz", hash = "sha256:87022b9dcb7593a601f7a40ca698227ac3ccb776b58cb7b06b8dc7f510995c34" }
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/08/46/9b217343648b366eb93447f5d93116e09a61956005794aed5ef95a2e9e2e/alibabacloud_openapi_util-0.2.4-py3-none-any.whl", hash = "sha256:a2474f230b5965ae9a8c286e0dc86132a887928d02d20b8182656cf6b1b6c5bd" },
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
]
[[package]]
name = "alibabacloud-tea"
version = "0.4.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
dependencies = [
{ name = "aiohttp" },
{ name = "requests" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/9a/7d/b22cb9a0d4f396ee0f3f9d7f26b76b9ed93d4101add7867a2c87ed2534f5/alibabacloud-tea-0.4.3.tar.gz", hash = "sha256:ec8053d0aa8d43ebe1deb632d5c5404339b39ec9a18a0707d57765838418504a" }
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
[[package]]
name = "alibabacloud-tea-openapi"
fix: upgrade crawl4ai to 0.8.0 (CVE-2026-26217) (#15415) ## Summary Upgrade crawl4ai from 0.7.6 to 0.8.0 to fix CVE-2026-26217. ## Vulnerability | Field | Value | |-------|-------| | **ID** | CVE-2026-26217 | | **Severity** | CRITICAL | | **Scanner** | trivy | | **Rule** | `CVE-2026-26217` | | **File** | `uv.lock` | | **Assessment** | Likely exploitable | **Description**: Crawl4AI Has Local File Inclusion in Docker API via file:// URLs ## Evidence **Scanner confirmation**: trivy rule `CVE-2026-26217` flagged this pattern. **Production code**: This file is in the production codebase, not test-only code. ## Threat Model Context This is a web service - vulnerabilities in request handlers are directly exploitable by remote attackers. ## Changes - `pyproject.toml` - `uv.lock` ## Verification - [x] Build passes - [x] Scanner re-scan confirms fix - [x] LLM code review passed --- *This change addresses a pattern flagged by static analysis. The code path handles user-influenced input and the fix reduces the attack surface against both manual and automated exploitation.* --- *Automated security fix by [OrbisAI Security](https://orbisappsec.com)*
2026-05-29 19:08:41 +05:30
version = "0.4.4"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
dependencies = [
{ name = "alibabacloud-credentials" },
{ name = "alibabacloud-gateway-spi" },
{ name = "alibabacloud-tea-util" },
{ name = "cryptography" },
{ name = "darabonba-core" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/30/93/138bcdc8fc596add73e37cf2073798f285284d1240bda9ee02f9384fc6be/alibabacloud_tea_openapi-0.4.4.tar.gz", hash = "sha256:1b0917bc03cd49417da64945e92731716d53e2eb8707b235f54e45b7473221ce" }
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/f5/5a/6bfc4506438c1809c486f66217ad11eab78157192b3d5707b4e2f4212f6c/alibabacloud_tea_openapi-0.4.4-py3-none-any.whl", hash = "sha256:cea6bc1fe35b0319a8752cb99eb0ecb0dab7ca1a71b99c12970ba0867410995f" },
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
]
[[package]]
name = "alibabacloud-tea-util"
version = "0.3.14"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
dependencies = [
{ name = "alibabacloud-tea" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/e9/ee/ea90be94ad781a5055db29556744681fc71190ef444ae53adba45e1be5f3/alibabacloud_tea_util-0.3.14.tar.gz", hash = "sha256:708e7c9f64641a3c9e0e566365d2f23675f8d7c2a3e2971d9402ceede0408cdb" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/72/9e/c394b4e2104766fb28a1e44e3ed36e4c7773b4d05c868e482be99d5635c9/alibabacloud_tea_util-0.3.14-py3-none-any.whl", hash = "sha256:10d3e5c340d8f7ec69dd27345eb2fc5a1dab07875742525edf07bbe86db93bfe" },
]
[[package]]
name = "aliyun-python-sdk-core"
version = "2.16.0"
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
dependencies = [
{ name = "cryptography" },
{ name = "jmespath" },
]
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/3e/09/da9f58eb38b4fdb97ba6523274fbf445ef6a06be64b433693da8307b4bec/aliyun-python-sdk-core-2.16.0.tar.gz", hash = "sha256:651caad597eb39d4fad6cf85133dffe92837d53bdf62db9d8f37dab6508bb8f9" }
[[package]]
name = "aliyun-python-sdk-kms"
version = "2.16.5"
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
dependencies = [
{ name = "aliyun-python-sdk-core" },
]
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a8/2c/9877d0e6b18ecf246df671ac65a5d1d9fecbf85bdcb5d43efbde0d4662eb/aliyun-python-sdk-kms-2.16.5.tar.gz", hash = "sha256:f328a8a19d83ecbb965ffce0ec1e9930755216d104638cd95ecd362753b813b3" }
Security: do not use litellm 1.82.7 and 1.82.8 (#13768) ### What problem does this PR solve? See [issue](https://github.com/BerriAI/litellm/issues/24518) from Litellm. Upgraded from `1.81.15` to `1.82.6`, so RAGFlow is safe as always. ### Type of change - [x] Security Co-authored-by: Jin Hai <haijin.chn@gmail.com> Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
2026-03-25 22:39:33 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/11/5c/0132193d7da2c735669a1ed103b142fd63c9455984d48c5a88a1a516efaa/aliyun_python_sdk_kms-2.16.5-py2.py3-none-any.whl", hash = "sha256:24b6cdc4fd161d2942619479c8d050c63ea9cd22b044fe33b60bbb60153786f0" },
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
]
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
[[package]]
name = "alphashape"
version = "1.3.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
dependencies = [
{ name = "click" },
{ name = "click-log" },
{ name = "networkx" },
{ name = "numpy" },
{ name = "rtree" },
{ name = "scipy" },
{ name = "shapely" },
{ name = "trimesh" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/2e/83/67ff905694df5b34a777123b59fdfd05998d5a31766f188aafbf5b340055/alphashape-1.3.1.tar.gz", hash = "sha256:7a27340afc5f8ed301577acec46bb0cf2bada5410045f7289142e735ef6977ec" }
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/e4/ad/77fad9d6f974ec58d837cb49fb9b483d6227a420c4f908c3578633de1d47/alphashape-1.3.1-py2.py3-none-any.whl", hash = "sha256:96a5ddd5f09534a35f03a8916aeeaac00fe4d6bec2f9ad78f87f57be3007f795" },
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "annotated-types"
version = "0.7.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ee/67/531ea369ba64dcff5ec9c3402f9f51bf748cec26dde048a2f973a4eea7f5/annotated_types-0.7.0.tar.gz", hash = "sha256:aff07c09a53a08bc8cfccb9c85b05f1aa9a2a6f23728d790723543408344ce89" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/78/b6/6307fbef88d9b5ee7421e68d78a9f162e0da4900bc5f5793f6d3d0e34fb8/annotated_types-0.7.0-py3-none-any.whl", hash = "sha256:1f02e8b43a8fbbc3f3e0d4f0f4bfc8131bcb4eebe8849b8e5c773f3a1c582a53" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "anthropic"
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
version = "0.76.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "anyio" },
{ name = "distro" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
{ name = "docstring-parser" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "httpx" },
{ name = "jiter" },
{ name = "pydantic" },
{ name = "sniffio" },
{ name = "typing-extensions" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/6e/be/d11abafaa15d6304826438170f7574d750218f49a106c54424a40cef4494/anthropic-0.76.0.tar.gz", hash = "sha256:e0cae6a368986d5cf6df743dfbb1b9519e6a9eee9c6c942ad8121c0b34416ffe" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/e5/70/7b0fd9c1a738f59d3babe2b4212031c34ab7d0fda4ffef15b58a55c5bcea/anthropic-0.76.0-py3-none-any.whl", hash = "sha256:81efa3113901192af2f0fe977d3ec73fdadb1e691586306c4256cd6d5ccc331c" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "anyio"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "4.12.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "idna" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/96/f0/5eb65b2bb0d09ac6776f2eb54adee6abe8228ea05b20a5ad0e4945de8aac/anyio-4.12.1.tar.gz", hash = "sha256:41cfcc3a4c85d3f05c932da7c26d0201ac36f72abd4435ba90d0464a3ffed703" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/38/0e/27be9fdef66e72d64c0cdc3cc2823101b80585f8119b5c112c2e8f5f7dab/anyio-4.12.1-py3-none-any.whl", hash = "sha256:d405828884fc140aa80a3c667b8beed277f1dfedec42ba031bd6ac3db606ab6c" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "anytree"
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
version = "2.13.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/bc/a8/eb55fab589c56f9b6be2b3fd6997aa04bb6f3da93b01154ce6fc8e799db2/anytree-2.13.0.tar.gz", hash = "sha256:c9d3aa6825fdd06af7ebb05b4ef291d2db63e62bb1f9b7d9b71354be9d362714" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/7b/98/f6aa7fe0783e42be3093d8ef1b0ecdc22c34c0d69640dfb37f56925cb141/anytree-2.13.0-py3-none-any.whl", hash = "sha256:4cbcf10df36b1f1cba131b7e487ff3edafc9d6e932a3c70071b5b768bab901ff" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Bump python to >=3.12 (#11846) ### What problem does this PR solve? Bump python to >=3.12 ### Type of change - [x] Refactoring
2025-12-09 19:55:25 +08:00
[[package]]
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
name = "apscheduler"
version = "3.11.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
dependencies = [
{ name = "tzlocal" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/07/12/3e4389e5920b4c1763390c6d371162f3784f86f85cd6d6c1bfe68eef14e2/apscheduler-3.11.2.tar.gz", hash = "sha256:2a9966b052ec805f020c8c4c3ae6e6a06e24b1bf19f2e11d91d8cca0473eef41" }
Bump python to >=3.12 (#11846) ### What problem does this PR solve? Bump python to >=3.12 ### Type of change - [x] Refactoring
2025-12-09 19:55:25 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/9f/64/2e54428beba8d9992aa478bb8f6de9e4ecaa5f8f513bcfd567ed7fb0262d/apscheduler-3.11.2-py3-none-any.whl", hash = "sha256:ce005177f741409db4e4dd40a7431b76feb856b9dd69d57e0da49d6715bfd26d" },
Bump python to >=3.12 (#11846) ### What problem does this PR solve? Bump python to >=3.12 ### Type of change - [x] Refactoring
2025-12-09 19:55:25 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "argon2-cffi"
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
version = "25.1.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "argon2-cffi-bindings" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/0e/89/ce5af8a7d472a67cc819d5d998aa8c82c5d860608c4db9f46f1162d7dab9/argon2_cffi-25.1.0.tar.gz", hash = "sha256:694ae5cc8a42f4c4e2bf2ca0e64e51e23a040c6a517a85074683d3959e1346c1" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/4f/d3/a8b22fa575b297cd6e3e3b0155c7e25db170edf1c74783d6a31a2490b8d9/argon2_cffi-25.1.0-py3-none-any.whl", hash = "sha256:fdc8b074db390fccb6eb4a3604ae7231f219aa669a2652e0f20e16ba513d5741" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "argon2-cffi-bindings"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "25.1.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "cffi" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/5c/2d/db8af0df73c1cf454f71b2bbe5e356b8c1f8041c979f505b3d3186e520a9/argon2_cffi_bindings-25.1.0.tar.gz", hash = "sha256:b957f3e6ea4d55d820e40ff76f450952807013d361a65d7f28acc0acbf29229d" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/1d/57/96b8b9f93166147826da5f90376e784a10582dd39a393c99bb62cfcf52f0/argon2_cffi_bindings-25.1.0-cp39-abi3-macosx_10_9_universal2.whl", hash = "sha256:aecba1723ae35330a008418a91ea6cfcedf6d31e5fbaa056a166462ff066d500" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0a/08/a9bebdb2e0e602dde230bdde8021b29f71f7841bd54801bcfd514acb5dcf/argon2_cffi_bindings-25.1.0-cp39-abi3-macosx_10_9_x86_64.whl", hash = "sha256:2630b6240b495dfab90aebe159ff784d08ea999aa4b0d17efa734055a07d2f44" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b6/02/d297943bcacf05e4f2a94ab6f462831dc20158614e5d067c35d4e63b9acb/argon2_cffi_bindings-25.1.0-cp39-abi3-macosx_11_0_arm64.whl", hash = "sha256:7aef0c91e2c0fbca6fc68e7555aa60ef7008a739cbe045541e438373bc54d2b0" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c1/93/44365f3d75053e53893ec6d733e4a5e3147502663554b4d864587c7828a7/argon2_cffi_bindings-25.1.0-cp39-abi3-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:1e021e87faa76ae0d413b619fe2b65ab9a037f24c60a1e6cc43457ae20de6dc6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/09/52/94108adfdd6e2ddf58be64f959a0b9c7d4ef2fa71086c38356d22dc501ea/argon2_cffi_bindings-25.1.0-cp39-abi3-manylinux_2_26_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:d3e924cfc503018a714f94a49a149fdc0b644eaead5d1f089330399134fa028a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/72/70/7a2993a12b0ffa2a9271259b79cc616e2389ed1a4d93842fac5a1f923ffd/argon2_cffi_bindings-25.1.0-cp39-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:c87b72589133f0346a1cb8d5ecca4b933e3c9b64656c9d175270a000e73b288d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/78/9a/4e5157d893ffc712b74dbd868c7f62365618266982b64accab26bab01edc/argon2_cffi_bindings-25.1.0-cp39-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:1db89609c06afa1a214a69a462ea741cf735b29a57530478c06eb81dd403de99" },
{ url = "https://mirrors.aliyun.com/pypi/packages/74/cd/15777dfde1c29d96de7f18edf4cc94c385646852e7c7b0320aa91ccca583/argon2_cffi_bindings-25.1.0-cp39-abi3-win32.whl", hash = "sha256:473bcb5f82924b1becbb637b63303ec8d10e84c8d241119419897a26116515d2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e2/c6/a759ece8f1829d1f162261226fbfd2c6832b3ff7657384045286d2afa384/argon2_cffi_bindings-25.1.0-cp39-abi3-win_amd64.whl", hash = "sha256:a98cd7d17e9f7ce244c0803cad3c23a7d379c301ba618a5fa76a67d116618b98" },
{ url = "https://mirrors.aliyun.com/pypi/packages/42/b9/f8d6fa329ab25128b7e98fd83a3cb34d9db5b059a9847eddb840a0af45dd/argon2_cffi_bindings-25.1.0-cp39-abi3-win_arm64.whl", hash = "sha256:b0fdbcf513833809c882823f98dc2f931cf659d9a1429616ac3adebb49f5db94" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "arrow"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "1.4.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "python-dateutil" },
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
{ name = "tzdata" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/b9/33/032cdc44182491aa708d06a68b62434140d8c50820a087fac7af37703357/arrow-1.4.0.tar.gz", hash = "sha256:ed0cc050e98001b8779e84d461b0098c4ac597e88704a655582b21d116e526d7" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ed/c9/d7977eaacb9df673210491da99e6a247e93df98c715fc43fd136ce1d3d33/arrow-1.4.0-py3-none-any.whl", hash = "sha256:749f0769958ebdc79c173ff0b0670d59051a535fa26e8eba02953dc19eb43205" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "arxiv"
version = "2.1.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "feedparser" },
{ name = "requests" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/fe/59/fe41f54bdfed776c2e9bcd6289e4c71349eb938241d89b4c97d0f33e8013/arxiv-2.1.3.tar.gz", hash = "sha256:32365221994d2cf05657c1fadf63a26efc8ccdec18590281ee03515bfef8bc4e" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/b7/7b/7bf42178d227b26d3daf94cdd22a72a4ed5bf235548c4f5aea49c51c6458/arxiv-2.1.3-py3-none-any.whl", hash = "sha256:6f43673ab770a9e848d7d4fc1894824df55edeac3c3572ea280c9ba2e3c0f39f" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: Add Asana data source integration and configuration options (#12239) ### What problem does this PR solve? change: Add Asana data source integration and configuration options ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2025-12-29 13:28:37 +08:00
[[package]]
name = "asana"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "5.2.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Add Asana data source integration and configuration options (#12239) ### What problem does this PR solve? change: Add Asana data source integration and configuration options ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2025-12-29 13:28:37 +08:00
dependencies = [
{ name = "certifi" },
{ name = "python-dateutil" },
{ name = "six" },
{ name = "urllib3" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/db/59/af14efdd03d332c33d4a77aed8f1f7151e3de5c2441e4bea3b1c6dbcc9d7/asana-5.2.2.tar.gz", hash = "sha256:d280ce2e8edf0355ccf21e548d887617ca8c926e1cb41309b8a173ca3181632c" }
Feat: Add Asana data source integration and configuration options (#12239) ### What problem does this PR solve? change: Add Asana data source integration and configuration options ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2025-12-29 13:28:37 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/26/5e/337125441af40aba86b087dee3dbe829413b6e42eac74defae2076926dbe/asana-5.2.2-py3-none-any.whl", hash = "sha256:1c8d15949a6cb9aa12363a5b7cfc6c0544cb3ae77290dd2e3255c0ec70668458" },
Feat: Add Asana data source integration and configuration options (#12239) ### What problem does this PR solve? change: Add Asana data source integration and configuration options ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2025-12-29 13:28:37 +08:00
]
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
[[package]]
name = "atlassian-python-api"
version = "4.0.7"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
dependencies = [
{ name = "beautifulsoup4" },
{ name = "deprecated" },
{ name = "jmespath" },
{ name = "oauthlib" },
{ name = "requests" },
{ name = "requests-oauthlib" },
{ name = "typing-extensions" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/40/e8/f23b7273e410c6fe9f98f9db25268c6736572f22a9566d1dc9ed3614bb68/atlassian_python_api-4.0.7.tar.gz", hash = "sha256:8d9cc6068b1d2a48eb434e22e57f6bbd918a47fac9e46b95b7a3cefb00fceacb" }
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/1d/83/e4f9976ce3c933a079b8931325e7a9c0a8bba7030a2cb85764c0048f3479/atlassian_python_api-4.0.7-py3-none-any.whl", hash = "sha256:46a70cb29eaab87c0a1697fccd3e25df1aa477e6aa4fb9ba936a9d46b425933c" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "attrs"
feat: bump Python minimum from 3.12 to 3.13, drop strenum backport (#14767) Closes #14753 ## What changed | File | Change | |---|---| | `pyproject.toml` | `requires-python` → `>=3.13,<3.15`; remove `strenum==0.4.15` | | `Dockerfile` | `uv python install 3.13`, `uv sync --python 3.13` | | `.github/workflows/tests.yml` | `uv sync --python 3.13` on both matrix legs | | `CLAUDE.md` | dev setup command + requirements note updated | | `deepdoc/parser/mineru_parser.py` | `from strenum import StrEnum` → `from enum import StrEnum` | | `agent/tools/code_exec.py` | same | `StrEnum` has been in the stdlib since Python 3.11 — the `strenum` backport package is no longer needed once the floor is 3.13. ## Why uv.lock is not regenerated `uv lock --python 3.13` fails because: 1. The infiniflow/graspologic fork pins `numpy>=1.26.4,<2.0.0` 2. `tensorflow-cpu>=2.20.0` (the first release with cp313 wheels) depends on `ml-dtypes>=0.5.1`, which requires `numpy>=2.1.0` 3. These two constraints are irreconcilable on Python 3.13 The lockfile regeneration requires loosening the `numpy` upper bound in the `infiniflow/graspologic` fork. Once that fork commit is updated and the SHA in `pyproject.toml:49` is bumped, `uv lock --python 3.13` will succeed. ## RFC corrections Two claims in the original RFC (#14753) did not hold up under code review: - **"graspologic hard-blocks 3.13"** — the infiniflow fork at the pinned commit has no `<3.13` Python constraint. The blocker is the transitive `numpy<2.0.0` conflict with tensorflow-cpu's test dependency, not a direct Python version cap. - **"free-threading throughput gains for I/O-bound workload"** — Python 3.13 free-threading requires a special `--disable-gil` build and provides no benefit for async I/O code (the GIL is already released during I/O). The real motivation is forward compatibility and improved error messages.
2026-05-15 08:40:53 +02:00
version = "26.1.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/9a/8e/82a0fe20a541c03148528be8cac2408564a6c9a0cc7e9171802bc1d26985/attrs-26.1.0.tar.gz", hash = "sha256:d03ceb89cb322a8fd706d4fb91940737b6642aa36998fe130a9bc96c985eff32" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/64/b4/17d4b0b2a2dc85a6df63d1157e028ed19f90d4cd97c36717afef2bc2f395/attrs-26.1.0-py3-none-any.whl", hash = "sha256:c647aa4a12dfbad9333ca4e71fe62ddc36f4e63b2d260a37a8b83d2f043ac309" },
feat: bump Python minimum from 3.12 to 3.13, drop strenum backport (#14767) Closes #14753 ## What changed | File | Change | |---|---| | `pyproject.toml` | `requires-python` → `>=3.13,<3.15`; remove `strenum==0.4.15` | | `Dockerfile` | `uv python install 3.13`, `uv sync --python 3.13` | | `.github/workflows/tests.yml` | `uv sync --python 3.13` on both matrix legs | | `CLAUDE.md` | dev setup command + requirements note updated | | `deepdoc/parser/mineru_parser.py` | `from strenum import StrEnum` → `from enum import StrEnum` | | `agent/tools/code_exec.py` | same | `StrEnum` has been in the stdlib since Python 3.11 — the `strenum` backport package is no longer needed once the floor is 3.13. ## Why uv.lock is not regenerated `uv lock --python 3.13` fails because: 1. The infiniflow/graspologic fork pins `numpy>=1.26.4,<2.0.0` 2. `tensorflow-cpu>=2.20.0` (the first release with cp313 wheels) depends on `ml-dtypes>=0.5.1`, which requires `numpy>=2.1.0` 3. These two constraints are irreconcilable on Python 3.13 The lockfile regeneration requires loosening the `numpy` upper bound in the `infiniflow/graspologic` fork. Once that fork commit is updated and the SHA in `pyproject.toml:49` is bumped, `uv lock --python 3.13` will succeed. ## RFC corrections Two claims in the original RFC (#14753) did not hold up under code review: - **"graspologic hard-blocks 3.13"** — the infiniflow fork at the pinned commit has no `<3.13` Python constraint. The blocker is the transitive `numpy<2.0.0` conflict with tensorflow-cpu's test dependency, not a direct Python version cap. - **"free-threading throughput gains for I/O-bound workload"** — Python 3.13 free-threading requires a special `--disable-gil` build and provides no benefit for async I/O code (the GIL is already released during I/O). The real motivation is forward compatibility and improved error messages.
2026-05-15 08:40:53 +02:00
]
[[package]]
name = "audioop-lts"
version = "0.2.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/38/53/946db57842a50b2da2e0c1e34bd37f36f5aadba1a929a3971c5d7841dbca/audioop_lts-0.2.2.tar.gz", hash = "sha256:64d0c62d88e67b98a1a5e71987b7aa7b5bcffc7dcee65b635823dbdd0a8dbbd0" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/de/d4/94d277ca941de5a507b07f0b592f199c22454eeaec8f008a286b3fbbacd6/audioop_lts-0.2.2-cp313-abi3-macosx_10_13_universal2.whl", hash = "sha256:fd3d4602dc64914d462924a08c1a9816435a2155d74f325853c1f1ac3b2d9800" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f8/5a/656d1c2da4b555920ce4177167bfeb8623d98765594af59702c8873f60ec/audioop_lts-0.2.2-cp313-abi3-macosx_10_13_x86_64.whl", hash = "sha256:550c114a8df0aafe9a05442a1162dfc8fec37e9af1d625ae6060fed6e756f303" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1b/83/ea581e364ce7b0d41456fb79d6ee0ad482beda61faf0cab20cbd4c63a541/audioop_lts-0.2.2-cp313-abi3-macosx_11_0_arm64.whl", hash = "sha256:9a13dc409f2564de15dd68be65b462ba0dde01b19663720c68c1140c782d1d75" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b8/3b/e8964210b5e216e5041593b7d33e97ee65967f17c282e8510d19c666dab4/audioop_lts-0.2.2-cp313-abi3-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:51c916108c56aa6e426ce611946f901badac950ee2ddaf302b7ed35d9958970d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c7/2e/0a1c52faf10d51def20531a59ce4c706cb7952323b11709e10de324d6493/audioop_lts-0.2.2-cp313-abi3-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:47eba38322370347b1c47024defbd36374a211e8dd5b0dcbce7b34fdb6f8847b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/75/e8/cd95eef479656cb75ab05dfece8c1f8c395d17a7c651d88f8e6e291a63ab/audioop_lts-0.2.2-cp313-abi3-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:ba7c3a7e5f23e215cb271516197030c32aef2e754252c4c70a50aaff7031a2c8" },
{ url = "https://mirrors.aliyun.com/pypi/packages/5c/1e/a0c42570b74f83efa5cca34905b3eef03f7ab09fe5637015df538a7f3345/audioop_lts-0.2.2-cp313-abi3-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:def246fe9e180626731b26e89816e79aae2276f825420a07b4a647abaa84becc" },
{ url = "https://mirrors.aliyun.com/pypi/packages/50/d5/8a0ae607ca07dbb34027bac8db805498ee7bfecc05fd2c148cc1ed7646e7/audioop_lts-0.2.2-cp313-abi3-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:e160bf9df356d841bb6c180eeeea1834085464626dc1b68fa4e1d59070affdc3" },
{ url = "https://mirrors.aliyun.com/pypi/packages/12/17/0d28c46179e7910bfb0bb62760ccb33edb5de973052cb2230b662c14ca2e/audioop_lts-0.2.2-cp313-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:4b4cd51a57b698b2d06cb9993b7ac8dfe89a3b2878e96bc7948e9f19ff51dba6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/84/ba/bd5d3806641564f2024e97ca98ea8f8811d4e01d9b9f9831474bc9e14f9e/audioop_lts-0.2.2-cp313-abi3-musllinux_1_2_ppc64le.whl", hash = "sha256:4a53aa7c16a60a6857e6b0b165261436396ef7293f8b5c9c828a3a203147ed4a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f9/5e/435ce8d5642f1f7679540d1e73c1c42d933331c0976eb397d1717d7f01a3/audioop_lts-0.2.2-cp313-abi3-musllinux_1_2_riscv64.whl", hash = "sha256:3fc38008969796f0f689f1453722a0f463da1b8a6fbee11987830bfbb664f623" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ae/3b/b909e76b606cbfd53875693ec8c156e93e15a1366a012f0b7e4fb52d3c34/audioop_lts-0.2.2-cp313-abi3-musllinux_1_2_s390x.whl", hash = "sha256:15ab25dd3e620790f40e9ead897f91e79c0d3ce65fe193c8ed6c26cffdd24be7" },
{ url = "https://mirrors.aliyun.com/pypi/packages/30/e7/8f1603b4572d79b775f2140d7952f200f5e6c62904585d08a01f0a70393a/audioop_lts-0.2.2-cp313-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:03f061a1915538fd96272bac9551841859dbb2e3bf73ebe4a23ef043766f5449" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b5/96/c37846df657ccdda62ba1ae2b6534fa90e2e1b1742ca8dcf8ebd38c53801/audioop_lts-0.2.2-cp313-abi3-win32.whl", hash = "sha256:3bcddaaf6cc5935a300a8387c99f7a7fbbe212a11568ec6cf6e4bc458c048636" },
{ url = "https://mirrors.aliyun.com/pypi/packages/34/a5/9d78fdb5b844a83da8a71226c7bdae7cc638861085fff7a1d707cb4823fa/audioop_lts-0.2.2-cp313-abi3-win_amd64.whl", hash = "sha256:a2c2a947fae7d1062ef08c4e369e0ba2086049a5e598fda41122535557012e9e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/34/25/20d8fde083123e90c61b51afb547bb0ea7e77bab50d98c0ab243d02a0e43/audioop_lts-0.2.2-cp313-abi3-win_arm64.whl", hash = "sha256:5f93a5db13927a37d2d09637ccca4b2b6b48c19cd9eda7b17a2e9f77edee6a6f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/58/a7/0a764f77b5c4ac58dc13c01a580f5d32ae8c74c92020b961556a43e26d02/audioop_lts-0.2.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:73f80bf4cd5d2ca7814da30a120de1f9408ee0619cc75da87d0641273d202a09" },
{ url = "https://mirrors.aliyun.com/pypi/packages/aa/ed/ebebedde1a18848b085ad0fa54b66ceb95f1f94a3fc04f1cd1b5ccb0ed42/audioop_lts-0.2.2-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:106753a83a25ee4d6f473f2be6b0966fc1c9af7e0017192f5531a3e7463dce58" },
{ url = "https://mirrors.aliyun.com/pypi/packages/cb/6e/11ca8c21af79f15dbb1c7f8017952ee8c810c438ce4e2b25638dfef2b02c/audioop_lts-0.2.2-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:fbdd522624141e40948ab3e8cdae6e04c748d78710e9f0f8d4dae2750831de19" },
{ url = "https://mirrors.aliyun.com/pypi/packages/84/52/0022f93d56d85eec5da6b9da6a958a1ef09e80c39f2cc0a590c6af81dcbb/audioop_lts-0.2.2-cp313-cp313t-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:143fad0311e8209ece30a8dbddab3b65ab419cbe8c0dde6e8828da25999be911" },
{ url = "https://mirrors.aliyun.com/pypi/packages/87/1d/48a889855e67be8718adbc7a01f3c01d5743c325453a5e81cf3717664aad/audioop_lts-0.2.2-cp313-cp313t-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:dfbbc74ec68a0fd08cfec1f4b5e8cca3d3cd7de5501b01c4b5d209995033cde9" },
{ url = "https://mirrors.aliyun.com/pypi/packages/98/a6/94b7213190e8077547ffae75e13ed05edc488653c85aa5c41472c297d295/audioop_lts-0.2.2-cp313-cp313t-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:cfcac6aa6f42397471e4943e0feb2244549db5c5d01efcd02725b96af417f3fe" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e9/e9/78450d7cb921ede0cfc33426d3a8023a3bda755883c95c868ee36db8d48d/audioop_lts-0.2.2-cp313-cp313t-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:752d76472d9804ac60f0078c79cdae8b956f293177acd2316cd1e15149aee132" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4f/e2/cd5439aad4f3e34ae1ee852025dc6aa8f67a82b97641e390bf7bd9891d3e/audioop_lts-0.2.2-cp313-cp313t-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:83c381767e2cc10e93e40281a04852facc4cd9334550e0f392f72d1c0a9c5753" },
{ url = "https://mirrors.aliyun.com/pypi/packages/68/4b/9d853e9076c43ebba0d411e8d2aa19061083349ac695a7d082540bad64d0/audioop_lts-0.2.2-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:c0022283e9556e0f3643b7c3c03f05063ca72b3063291834cca43234f20c60bb" },
{ url = "https://mirrors.aliyun.com/pypi/packages/58/26/4bae7f9d2f116ed5593989d0e521d679b0d583973d203384679323d8fa85/audioop_lts-0.2.2-cp313-cp313t-musllinux_1_2_ppc64le.whl", hash = "sha256:a2d4f1513d63c795e82948e1305f31a6d530626e5f9f2605408b300ae6095093" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b2/67/a9f4fb3e250dda9e9046f8866e9fa7d52664f8985e445c6b4ad6dfb55641/audioop_lts-0.2.2-cp313-cp313t-musllinux_1_2_riscv64.whl", hash = "sha256:c9c8e68d8b4a56fda8c025e538e639f8c5953f5073886b596c93ec9b620055e7" },
{ url = "https://mirrors.aliyun.com/pypi/packages/70/f7/3de86562db0121956148bcb0fe5b506615e3bcf6e63c4357a612b910765a/audioop_lts-0.2.2-cp313-cp313t-musllinux_1_2_s390x.whl", hash = "sha256:96f19de485a2925314f5020e85911fb447ff5fbef56e8c7c6927851b95533a1c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f1/32/fd772bf9078ae1001207d2df1eef3da05bea611a87dd0e8217989b2848fa/audioop_lts-0.2.2-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:e541c3ef484852ef36545f66209444c48b28661e864ccadb29daddb6a4b8e5f5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4f/41/affea7181592ab0ab560044632571a38edaf9130b84928177823fbf3176a/audioop_lts-0.2.2-cp313-cp313t-win32.whl", hash = "sha256:d5e73fa573e273e4f2e5ff96f9043858a5e9311e94ffefd88a3186a910c70917" },
{ url = "https://mirrors.aliyun.com/pypi/packages/28/2b/0372842877016641db8fc54d5c88596b542eec2f8f6c20a36fb6612bf9ee/audioop_lts-0.2.2-cp313-cp313t-win_amd64.whl", hash = "sha256:9191d68659eda01e448188f60364c7763a7ca6653ed3f87ebb165822153a8547" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ee/ca/baf2b9cc7e96c179bb4a54f30fcd83e6ecb340031bde68f486403f943768/audioop_lts-0.2.2-cp313-cp313t-win_arm64.whl", hash = "sha256:c174e322bb5783c099aaf87faeb240c8d210686b04bd61dfd05a8e5a83d88969" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
[[package]]
name = "authlib"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.7.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "cryptography" },
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
{ name = "joserfc" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/36/98/7d93f30d029643c0275dbc0bd6d5a6f670661ee6c9a94d93af7ab4887600/authlib-1.7.2.tar.gz", hash = "sha256:2cea25fefcd4e7173bdf1372c0afc265c8034b23a8cd5dcb6a9164b826c64231" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/fb/95/adcb68e20c34162e9135f370d6e31737719c2b6f94bc953fe7ed1f10fe21/authlib-1.7.2-py2.py3-none-any.whl", hash = "sha256:3e1faedc9d87e7d56a164eca3ccb6ace0d61b94abe83e92242f8dc8bba9b4a9f" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "autograd"
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
version = "1.8.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "numpy" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/67/1c/3c24ec03c8ba4decc742b1df5a10c52f98c84ca8797757f313e7bdcdf276/autograd-1.8.0.tar.gz", hash = "sha256:107374ded5b09fc8643ac925348c0369e7b0e73bbed9565ffd61b8fd04425683" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/84/ea/e16f0c423f7d83cf8b79cae9452040fb7b2e020c7439a167ee7c317de448/autograd-1.8.0-py3-none-any.whl", hash = "sha256:4ab9084294f814cf56c280adbe19612546a35574d67c574b04933c7d2ecb7d78" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "azure-core"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.37.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "requests" },
{ name = "typing-extensions" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ef/83/41c9371c8298999c67b007e308a0a3c4d6a59c6908fa9c62101f031f886f/azure_core-1.37.0.tar.gz", hash = "sha256:7064f2c11e4b97f340e8e8c6d923b822978be3016e46b7bc4aa4b337cfb48aee" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ee/34/a9914e676971a13d6cc671b1ed172f9804b50a3a80a143ff196e52f4c7ee/azure_core-1.37.0-py3-none-any.whl", hash = "sha256:b3abe2c59e7d6bb18b38c275a5029ff80f98990e7c90a5e646249a56630fcc19" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "azure-identity"
Fix: revert xgboost version to 1.6.0 (#13984) ### What problem does this PR solve? Revert xgboost version to 1.6.0 ### Type of change - [ ] Bug Fix (non-breaking change which fixes an issue) - [ ] New Feature (non-breaking change which adds functionality) - [ ] Documentation Update - [ ] Refactoring - [ ] Performance Improvement - [ ] Other (please describe): <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Chores** * Updated xgboost dependency from version 3.2.0 to 1.6.0 <!-- end of auto-generated comment: release notes by coderabbit.ai -->
2026-04-08 19:53:47 +08:00
version = "1.25.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "azure-core" },
{ name = "cryptography" },
{ name = "msal" },
{ name = "msal-extensions" },
{ name = "typing-extensions" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/c5/0e/3a63efb48aa4a5ae2cfca61ee152fbcb668092134d3eb8bfda472dd5c617/azure_identity-1.25.3.tar.gz", hash = "sha256:ab23c0d63015f50b630ef6c6cf395e7262f439ce06e5d07a64e874c724f8d9e6" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/49/9a/417b3a533e01953a7c618884df2cb05a71e7b68bdbce4fbdb62349d2a2e8/azure_identity-1.25.3-py3-none-any.whl", hash = "sha256:f4d0b956a8146f30333e071374171f3cfa7bdb8073adb8c3814b65567aa7447c" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "azure-storage-blob"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "12.28.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "azure-core" },
{ name = "cryptography" },
{ name = "isodate" },
{ name = "typing-extensions" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/71/24/072ba8e27b0e2d8fec401e9969b429d4f5fc4c8d4f0f05f4661e11f7234a/azure_storage_blob-12.28.0.tar.gz", hash = "sha256:e7d98ea108258d29aa0efbfd591b2e2075fa1722a2fae8699f0b3c9de11eff41" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/d8/3a/6ef2047a072e54e1142718d433d50e9514c999a58f51abfff7902f3a72f8/azure_storage_blob-12.28.0-py3-none-any.whl", hash = "sha256:00fb1db28bf6a7b7ecaa48e3b1d5c83bfadacc5a678b77826081304bd87d6461" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "azure-storage-file-datalake"
version = "12.16.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "azure-core" },
{ name = "azure-storage-blob" },
{ name = "isodate" },
{ name = "typing-extensions" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/71/a6/980d2d1405ae5397b618cc9a21b4530fb7e6c9078ccf48b5ce0eec1b25cd/azure-storage-file-datalake-12.16.0.tar.gz", hash = "sha256:3185580e4e438162ef84fb88cb46b2ef248dafbfb07f53297762417bb7000333" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/42/56/a30d062af3100b3ec3e515fc1f40d38979e8508bb962231530702309aa4b/azure_storage_file_datalake-12.16.0-py3-none-any.whl", hash = "sha256:da57ec6cf5640b92bbd0ba61478f51e67c63b94843fa748b3b6599f1adba5837" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "babel"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "2.17.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/7d/6b/d52e42361e1aa00709585ecc30b3f9684b3ab62530771402248b1b1d6240/babel-2.17.0.tar.gz", hash = "sha256:0c54cffb19f690cdcc52a3b50bcbf71e07a808d1c80d549f2459b9d2cf0afb9d" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/b7/b8/3fe70c75fe32afc4bb507f75563d39bc5642255d1d94f1f23604725780bf/babel-2.17.0-py3-none-any.whl", hash = "sha256:4d0b53093fdfb4b21c92b5213dba5a1b23885afa8383709427046b21c366e5f2" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: add basic Langfuse support for LLM module (#6443) ### What problem does this PR solve? #6155 Add basic Langfuse support for LLM module. A trace example: <img width="755" alt="image" src="https://github.com/user-attachments/assets/25c1f852-5116-486c-a47f-6097187142ca" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-03-24 13:18:47 +08:00
[[package]]
name = "backoff"
version = "2.2.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/47/d7/5bbeb12c44d7c4f2fb5b56abce497eb5ed9f34d85701de869acedd602619/backoff-2.2.1.tar.gz", hash = "sha256:03f829f5bb1923180821643f8753b0502c3b682293992485b0eef2807afa5cba" }
Feat: add basic Langfuse support for LLM module (#6443) ### What problem does this PR solve? #6155 Add basic Langfuse support for LLM module. A trace example: <img width="755" alt="image" src="https://github.com/user-attachments/assets/25c1f852-5116-486c-a47f-6097187142ca" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-03-24 13:18:47 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/df/73/b6e24bd22e6720ca8ee9a85a0c4a2971af8497d8f3193fa05390cbd46e09/backoff-2.2.1-py3-none-any.whl", hash = "sha256:63579f9a0628e06278f7e47b7d7d5b6ce20dc65c5e96a6f3ca99a6adca0396e8" },
Feat: add basic Langfuse support for LLM module (#6443) ### What problem does this PR solve? #6155 Add basic Langfuse support for LLM module. A trace example: <img width="755" alt="image" src="https://github.com/user-attachments/assets/25c1f852-5116-486c-a47f-6097187142ca" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-03-24 13:18:47 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "bce-python-sdk"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "0.9.59"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "future" },
{ name = "pycryptodome" },
{ name = "six" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/c1/8e/ddfacf065fd0a514bda38b489988ea21636ac3be09c79239f24cdc36d71b/bce_python_sdk-0.9.59.tar.gz", hash = "sha256:54ad09394b0a5baf8c8ef87ac919f9d111c1b0536086286b80ada71651d8e4c8" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/2f/b0/38ea413e3a4aa44c199ff74001b3b2510b6b0f237c7840237976094ab574/bce_python_sdk-0.9.59-py3-none-any.whl", hash = "sha256:9a63ffc36ac5cb984b79ce6909288f00862010eda576f7575c7f0fb7cdef419c" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: add local & ssh provider in admin panel (#15039) ### What problem does this PR solve? Feat: add local & ssh provider in admin panel ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-20 16:56:20 +08:00
[[package]]
name = "bcrypt"
version = "5.0.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/d4/36/3329e2518d70ad8e2e5817d5a4cac6bba05a47767ec416c7d020a965f408/bcrypt-5.0.0.tar.gz", hash = "sha256:f748f7c2d6fd375cc93d3fba7ef4a9e3a092421b8dbf34d8d4dc06be9492dfdd" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/13/85/3e65e01985fddf25b64ca67275bb5bdb4040bd1a53b66d355c6c37c8a680/bcrypt-5.0.0-cp313-cp313t-macosx_10_12_universal2.whl", hash = "sha256:f3c08197f3039bec79cee59a606d62b96b16669cff3949f21e74796b6e3cd2be" },
{ url = "https://mirrors.aliyun.com/pypi/packages/44/dc/01eb79f12b177017a726cbf78330eb0eb442fae0e7b3dfd84ea2849552f3/bcrypt-5.0.0-cp313-cp313t-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:200af71bc25f22006f4069060c88ed36f8aa4ff7f53e67ff04d2ab3f1e79a5b2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8c/cf/e82388ad5959c40d6afd94fb4743cc077129d45b952d46bdc3180310e2df/bcrypt-5.0.0-cp313-cp313t-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:baade0a5657654c2984468efb7d6c110db87ea63ef5a4b54732e7e337253e44f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ec/86/7134b9dae7cf0efa85671651341f6afa695857fae172615e960fb6a466fa/bcrypt-5.0.0-cp313-cp313t-manylinux_2_28_aarch64.whl", hash = "sha256:c58b56cdfb03202b3bcc9fd8daee8e8e9b6d7e3163aa97c631dfcfcc24d36c86" },
{ url = "https://mirrors.aliyun.com/pypi/packages/cc/82/6296688ac1b9e503d034e7d0614d56e80c5d1a08402ff856a4549cb59207/bcrypt-5.0.0-cp313-cp313t-manylinux_2_28_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:4bfd2a34de661f34d0bda43c3e4e79df586e4716ef401fe31ea39d69d581ef23" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d1/18/884a44aa47f2a3b88dd09bc05a1e40b57878ecd111d17e5bba6f09f8bb77/bcrypt-5.0.0-cp313-cp313t-manylinux_2_28_x86_64.whl", hash = "sha256:ed2e1365e31fc73f1825fa830f1c8f8917ca1b3ca6185773b349c20fd606cec2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0e/8f/371a3ab33c6982070b674f1788e05b656cfbf5685894acbfef0c65483a59/bcrypt-5.0.0-cp313-cp313t-manylinux_2_34_aarch64.whl", hash = "sha256:83e787d7a84dbbfba6f250dd7a5efd689e935f03dd83b0f919d39349e1f23f83" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b1/34/7e4e6abb7a8778db6422e88b1f06eb07c47682313997ee8a8f9352e5a6f1/bcrypt-5.0.0-cp313-cp313t-manylinux_2_34_x86_64.whl", hash = "sha256:137c5156524328a24b9fac1cb5db0ba618bc97d11970b39184c1d87dc4bf1746" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c0/1b/54f416be2499bd72123c70d98d36c6cd61a4e33d9b89562c22481c81bb30/bcrypt-5.0.0-cp313-cp313t-musllinux_1_1_aarch64.whl", hash = "sha256:38cac74101777a6a7d3b3e3cfefa57089b5ada650dce2baf0cbdd9d65db22a9e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/13/62/062c24c7bcf9d2826a1a843d0d605c65a755bc98002923d01fd61270705a/bcrypt-5.0.0-cp313-cp313t-musllinux_1_1_x86_64.whl", hash = "sha256:d8d65b564ec849643d9f7ea05c6d9f0cd7ca23bdd4ac0c2dbef1104ab504543d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d5/c8/1fdbfc8c0f20875b6b4020f3c7dc447b8de60aa0be5faaf009d24242aec9/bcrypt-5.0.0-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:741449132f64b3524e95cd30e5cd3343006ce146088f074f31ab26b94e6c75ba" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a6/c1/8b84545382d75bef226fbc6588af0f7b7d095f7cd6a670b42a86243183cd/bcrypt-5.0.0-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:212139484ab3207b1f0c00633d3be92fef3c5f0af17cad155679d03ff2ee1e41" },
{ url = "https://mirrors.aliyun.com/pypi/packages/10/a6/ffb49d4254ed085e62e3e5dd05982b4393e32fe1e49bb1130186617c29cd/bcrypt-5.0.0-cp313-cp313t-win32.whl", hash = "sha256:9d52ed507c2488eddd6a95bccee4e808d3234fa78dd370e24bac65a21212b861" },
{ url = "https://mirrors.aliyun.com/pypi/packages/48/a9/259559edc85258b6d5fc5471a62a3299a6aa37a6611a169756bf4689323c/bcrypt-5.0.0-cp313-cp313t-win_amd64.whl", hash = "sha256:f6984a24db30548fd39a44360532898c33528b74aedf81c26cf29c51ee47057e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2d/df/9714173403c7e8b245acf8e4be8876aac64a209d1b392af457c79e60492e/bcrypt-5.0.0-cp313-cp313t-win_arm64.whl", hash = "sha256:9fffdb387abe6aa775af36ef16f55e318dcda4194ddbf82007a6f21da29de8f5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/84/29/6237f151fbfe295fe3e074ecc6d44228faa1e842a81f6d34a02937ee1736/bcrypt-5.0.0-cp38-abi3-macosx_10_12_universal2.whl", hash = "sha256:fc746432b951e92b58317af8e0ca746efe93e66555f1b40888865ef5bf56446b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/45/b6/4c1205dde5e464ea3bd88e8742e19f899c16fa8916fb8510a851fae985b5/bcrypt-5.0.0-cp38-abi3-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:c2388ca94ffee269b6038d48747f4ce8df0ffbea43f31abfa18ac72f0218effb" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3b/71/427945e6ead72ccffe77894b2655b695ccf14ae1866cd977e185d606dd2f/bcrypt-5.0.0-cp38-abi3-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:560ddb6ec730386e7b3b26b8b4c88197aaed924430e7b74666a586ac997249ef" },
{ url = "https://mirrors.aliyun.com/pypi/packages/17/72/c344825e3b83c5389a369c8a8e58ffe1480b8a699f46c127c34580c4666b/bcrypt-5.0.0-cp38-abi3-manylinux_2_28_aarch64.whl", hash = "sha256:d79e5c65dcc9af213594d6f7f1fa2c98ad3fc10431e7aa53c176b441943efbdd" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0b/7e/d4e47d2df1641a36d1212e5c0514f5291e1a956a7749f1e595c07a972038/bcrypt-5.0.0-cp38-abi3-manylinux_2_28_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:2b732e7d388fa22d48920baa267ba5d97cca38070b69c0e2d37087b381c681fd" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0f/c3/0ae57a68be2039287ec28bc463b82e4b8dc23f9d12c0be331f4782e19108/bcrypt-5.0.0-cp38-abi3-manylinux_2_28_x86_64.whl", hash = "sha256:0c8e093ea2532601a6f686edbc2c6b2ec24131ff5c52f7610dd64fa4553b5464" },
{ url = "https://mirrors.aliyun.com/pypi/packages/45/2b/77424511adb11e6a99e3a00dcc7745034bee89036ad7d7e255a7e47be7d8/bcrypt-5.0.0-cp38-abi3-manylinux_2_34_aarch64.whl", hash = "sha256:5b1589f4839a0899c146e8892efe320c0fa096568abd9b95593efac50a87cb75" },
{ url = "https://mirrors.aliyun.com/pypi/packages/43/0a/405c753f6158e0f3f14b00b462d8bca31296f7ecfc8fc8bc7919c0c7d73a/bcrypt-5.0.0-cp38-abi3-manylinux_2_34_x86_64.whl", hash = "sha256:89042e61b5e808b67daf24a434d89bab164d4de1746b37a8d173b6b14f3db9ff" },
{ url = "https://mirrors.aliyun.com/pypi/packages/62/83/b3efc285d4aadc1fa83db385ec64dcfa1707e890eb42f03b127d66ac1b7b/bcrypt-5.0.0-cp38-abi3-musllinux_1_1_aarch64.whl", hash = "sha256:e3cf5b2560c7b5a142286f69bde914494b6d8f901aaa71e453078388a50881c4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/95/7d/47ee337dacecde6d234890fe929936cb03ebc4c3a7460854bbd9c97780b8/bcrypt-5.0.0-cp38-abi3-musllinux_1_1_x86_64.whl", hash = "sha256:f632fd56fc4e61564f78b46a2269153122db34988e78b6be8b32d28507b7eaeb" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d6/3a/43d494dfb728f55f4e1cf8fd435d50c16a2d75493225b54c8d06122523c6/bcrypt-5.0.0-cp38-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:801cad5ccb6b87d1b430f183269b94c24f248dddbbc5c1f78b6ed231743e001c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/55/ab/a0727a4547e383e2e22a630e0f908113db37904f58719dc48d4622139b5c/bcrypt-5.0.0-cp38-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:3cf67a804fc66fc217e6914a5635000259fbbbb12e78a99488e4d5ba445a71eb" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1b/bb/461f352fdca663524b4643d8b09e8435b4990f17fbf4fea6bc2a90aa0cc7/bcrypt-5.0.0-cp38-abi3-win32.whl", hash = "sha256:3abeb543874b2c0524ff40c57a4e14e5d3a66ff33fb423529c88f180fd756538" },
{ url = "https://mirrors.aliyun.com/pypi/packages/41/aa/4190e60921927b7056820291f56fc57d00d04757c8b316b2d3c0d1d6da2c/bcrypt-5.0.0-cp38-abi3-win_amd64.whl", hash = "sha256:35a77ec55b541e5e583eb3436ffbbf53b0ffa1fa16ca6782279daf95d146dcd9" },
{ url = "https://mirrors.aliyun.com/pypi/packages/54/12/cd77221719d0b39ac0b55dbd39358db1cd1246e0282e104366ebbfb8266a/bcrypt-5.0.0-cp38-abi3-win_arm64.whl", hash = "sha256:cde08734f12c6a4e28dc6755cd11d3bdfea608d93d958fffbe95a7026ebe4980" },
{ url = "https://mirrors.aliyun.com/pypi/packages/5d/ba/2af136406e1c3839aea9ecadc2f6be2bcd1eff255bd451dd39bcf302c47a/bcrypt-5.0.0-cp39-abi3-macosx_10_12_universal2.whl", hash = "sha256:0c418ca99fd47e9c59a301744d63328f17798b5947b0f791e9af3c1c499c2d0a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ac/ee/2f4985dbad090ace5ad1f7dd8ff94477fe089b5fab2040bd784a3d5f187b/bcrypt-5.0.0-cp39-abi3-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:ddb4e1500f6efdd402218ffe34d040a1196c072e07929b9820f363a1fd1f4191" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e4/6e/b77ade812672d15cf50842e167eead80ac3514f3beacac8902915417f8b7/bcrypt-5.0.0-cp39-abi3-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:7aeef54b60ceddb6f30ee3db090351ecf0d40ec6e2abf41430997407a46d2254" },
{ url = "https://mirrors.aliyun.com/pypi/packages/36/c4/ed00ed32f1040f7990dac7115f82273e3c03da1e1a1587a778d8cea496d8/bcrypt-5.0.0-cp39-abi3-manylinux_2_28_aarch64.whl", hash = "sha256:f0ce778135f60799d89c9693b9b398819d15f1921ba15fe719acb3178215a7db" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e7/c4/fa6e16145e145e87f1fa351bbd54b429354fd72145cd3d4e0c5157cf4c70/bcrypt-5.0.0-cp39-abi3-manylinux_2_28_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:a71f70ee269671460b37a449f5ff26982a6f2ba493b3eabdd687b4bf35f875ac" },
{ url = "https://mirrors.aliyun.com/pypi/packages/24/b4/11f8a31d8b67cca3371e046db49baa7c0594d71eb40ac8121e2fc0888db0/bcrypt-5.0.0-cp39-abi3-manylinux_2_28_x86_64.whl", hash = "sha256:f8429e1c410b4073944f03bd778a9e066e7fad723564a52ff91841d278dfc822" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ac/31/79f11865f8078e192847d2cb526e3fa27c200933c982c5b2869720fa5fce/bcrypt-5.0.0-cp39-abi3-manylinux_2_34_aarch64.whl", hash = "sha256:edfcdcedd0d0f05850c52ba3127b1fce70b9f89e0fe5ff16517df7e81fa3cbb8" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d4/8d/5e43d9584b3b3591a6f9b68f755a4da879a59712981ef5ad2a0ac1379f7a/bcrypt-5.0.0-cp39-abi3-manylinux_2_34_x86_64.whl", hash = "sha256:611f0a17aa4a25a69362dcc299fda5c8a3d4f160e2abb3831041feb77393a14a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/89/48/44590e3fc158620f680a978aafe8f87a4c4320da81ed11552f0323aa9a57/bcrypt-5.0.0-cp39-abi3-musllinux_1_1_aarch64.whl", hash = "sha256:db99dca3b1fdc3db87d7c57eac0c82281242d1eabf19dcb8a6b10eb29a2e72d1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/5f/85/e4fbfc46f14f47b0d20493669a625da5827d07e8a88ee460af6cd9768b44/bcrypt-5.0.0-cp39-abi3-musllinux_1_1_x86_64.whl", hash = "sha256:5feebf85a9cefda32966d8171f5db7e3ba964b77fdfe31919622256f80f9cf42" },
{ url = "https://mirrors.aliyun.com/pypi/packages/25/ae/479f81d3f4594456a01ea2f05b132a519eff9ab5768a70430fa1132384b1/bcrypt-5.0.0-cp39-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:3ca8a166b1140436e058298a34d88032ab62f15aae1c598580333dc21d27ef10" },
{ url = "https://mirrors.aliyun.com/pypi/packages/df/d2/36a086dee1473b14276cd6ea7f61aef3b2648710b5d7f1c9e032c29b859f/bcrypt-5.0.0-cp39-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:61afc381250c3182d9078551e3ac3a41da14154fbff647ddf52a769f588c4172" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c0/f6/688d2cd64bfd0b14d805ddb8a565e11ca1fb0fd6817175d58b10052b6d88/bcrypt-5.0.0-cp39-abi3-win32.whl", hash = "sha256:64d7ce196203e468c457c37ec22390f1a61c85c6f0b8160fd752940ccfb3a683" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9f/b9/9d9a641194a730bda138b3dfe53f584d61c58cd5230e37566e83ec2ffa0d/bcrypt-5.0.0-cp39-abi3-win_amd64.whl", hash = "sha256:64ee8434b0da054d830fa8e89e1c8bf30061d539044a39524ff7dec90481e5c2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/27/44/d2ef5e87509158ad2187f4dd0852df80695bb1ee0cfe0a684727b01a69e0/bcrypt-5.0.0-cp39-abi3-win_arm64.whl", hash = "sha256:f2347d3534e76bf50bca5500989d6c1d05ed64b440408057a37673282c654927" },
Feat: add local & ssh provider in admin panel (#15039) ### What problem does this PR solve? Feat: add local & ssh provider in admin panel ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-20 16:56:20 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "beartype"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "0.22.9"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/c7/94/1009e248bbfbab11397abca7193bea6626806be9a327d399810d523a07cb/beartype-0.22.9.tar.gz", hash = "sha256:8f82b54aa723a2848a56008d18875f91c1db02c32ef6a62319a002e3e25a975f" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/71/cc/18245721fa7747065ab478316c7fea7c74777d07f37ae60db2e84f8172e8/beartype-0.22.9-py3-none-any.whl", hash = "sha256:d16c9bbc61ea14637596c5f6fbff2ee99cbe3573e46a716401734ef50c3060c2" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "beautifulsoup4"
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
version = "4.13.5"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "soupsieve" },
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
{ name = "typing-extensions" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/85/2e/3e5079847e653b1f6dc647aa24549d68c6addb4c595cc0d902d1b19308ad/beautifulsoup4-4.13.5.tar.gz", hash = "sha256:5e70131382930e7c3de33450a2f54a63d5e4b19386eab43a5b34d594268f3695" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/04/eb/f4151e0c7377a6e08a38108609ba5cede57986802757848688aeedd1b9e8/beautifulsoup4-4.13.5-py3-none-any.whl", hash = "sha256:642085eaa22233aceadff9c69651bc51e8bf3f874fb6d7104ece2beb24b47c4a" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "bibtexparser"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.4.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "pyparsing" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/92/8d/e296c7af03757debd8fc80df2898cbed4fb69fc61ed2c9b4a1d42e923a9e/bibtexparser-1.4.3.tar.gz", hash = "sha256:a9c7ded64bc137720e4df0b1b7f12734edc1361185f1c9097048ff7c35af2b8f" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "bio"
version = "1.7.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "biopython" },
{ name = "gprofiler-official" },
{ name = "mygene" },
{ name = "pandas" },
{ name = "pooch" },
{ name = "requests" },
{ name = "tqdm" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/58/ba/fdaa4c286ed50f96835a5f81c72d6c76933fb890ee1ff2269b6110ea851e/bio-1.7.1.tar.gz", hash = "sha256:df3252905b0b1e739eca3760c91fd519d5af07b09632df25c2bd4ecd20da2724" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/cb/40/747f3038ac636e520da52f7b9f5721779a50f88fdfc165847b0d8127dae2/bio-1.7.1-py3-none-any.whl", hash = "sha256:851545804b08413a3f27fd5131edefc30acfdee513919eebabb29678d8632218" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "biopython"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "1.86"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "numpy" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/9d/61/c59a849bd457c8a1b408ae828dbcc15e674962b5a29705e869e15b32bf25/biopython-1.86.tar.gz", hash = "sha256:93a50b586a4d2cec68ab2f99d03ef583c5761d8fba5535cb8e81da781d0d92ff" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/fd/6e/84d6c66ab93095aa7adb998a8eef045328470eafd36b9237c4db213e587c/biopython-1.86-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:fb3a11a98e49428720dca227e2a5bdd57c973ee7c4df3cf6734c0aa13fd134c7" },
{ url = "https://mirrors.aliyun.com/pypi/packages/12/75/60386f2640f13765b1651f2f26d8b4f893c46ee663df3ca76eda966d4f6a/biopython-1.86-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:e161f3d3b6e65fbfd1ce22a01c3e9fa9da789adde4972fd0cc2370795ea5357b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/dd/de/a39adb98a0552a257219503c236ef17f007598af55326c0d143db52e5a92/biopython-1.86-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:5aa8c9e92ee6fe59dfe0d2c2daf9a9eec6b812c78328caad038f79163c500218" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0b/c7/b2e7aca3de8981f4ecb6ab1e0334c3c4a512e5e9898b57b3d8734b086da7/biopython-1.86-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:593ec6a2a4fedec08ddcee1a8a0e0b0ed56835b2714904b352ec4a93d5b9d973" },
{ url = "https://mirrors.aliyun.com/pypi/packages/52/ed/e6647b0b9cf2bb67347612e8e443b84378c44768a8d8439276e4ba881178/biopython-1.86-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:dd2f9ebf9b14d67ca92f48779c4f0ba404c35dba3e8b9d6c34d1a3591c3b746d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ff/37/f6a14b835842c66a52f212136a99416265f5ce76813d668ceac1cb306357/biopython-1.86-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:137fe9aafd93baa5127d17534b473f6646f92a883f52b34f7c306b800ac50038" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f2/73/0eac930016c509763c174a0e25e92e6d7a711f6f5de1f7001e54fd5c49f7/biopython-1.86-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:e784dc8382430c9893aa084ca18fe8a8815b5811f1c324492ef3f4b54e664fff" },
{ url = "https://mirrors.aliyun.com/pypi/packages/00/aa/26e836274d03402e8011b04a1714d4ac2f704add303a493e54d2d5646973/biopython-1.86-cp313-cp313-win32.whl", hash = "sha256:5329a777ba90ea624447173046e77c4df2862acc46eea4e94fe2211fe041750f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ae/27/fa1f8fa57f2ac8fdc41d14ab36001b8ba0fce5eac01585227b99a4da0e9d/biopython-1.86-cp313-cp313-win_amd64.whl", hash = "sha256:f6f2f1dc75423b15d8a22b8eceae32785736612b6740688526401b8c2d821270" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "biothings-client"
Light GraphRAG (#4585) ### What problem does this PR solve? #4543 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-01-22 19:43:14 +08:00
version = "0.4.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "httpx" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ef/2f/2ef9115e317f4c2acb690b341f40de751b16d14169fdbb8d6eb86964166c/biothings_client-0.4.1.tar.gz", hash = "sha256:5b34e09c905280b5bd2538f1f34b6fc780c53c8da9b4074e3ff304836046f613" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/dc/6d/130477cfbd7294949b919c45cc1ed14a642cec95afba06a54400a4419235/biothings_client-0.4.1-py3-none-any.whl", hash = "sha256:9cbc17461b2bf6af6ed200929b886d6670d450af2034b428cd833f725695265a" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "blinker"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "1.9.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/21/28/9b3f50ce0e048515135495f198351908d99540d69bfdc8c1d15b73dc55ce/blinker-1.9.0.tar.gz", hash = "sha256:b4ce2265a7abece45e7cc896e98dbebe6cead56bcf805a3d23136d145f5445bf" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/10/cb/f2ad4230dc2eb1a74edf38f1a38b9b52277f75bef262d8908e60d957e13c/blinker-1.9.0-py3-none-any.whl", hash = "sha256:ba0efaa9080b619ff2f3459d1d500c57bddea4a6b424b60a91141db6fd2f08bc" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
[[package]]
name = "blis"
version = "1.3.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
dependencies = [
{ name = "numpy" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/d0/d0/d8cc8c9a4488a787e7fa430f6055e5bd1ddb22c340a751d9e901b82e2efe/blis-1.3.3.tar.gz", hash = "sha256:034d4560ff3cc43e8aa37e188451b0440e3261d989bb8a42ceee865607715ecd" }
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/e6/f7/d26e62d9be3d70473a63e0a5d30bae49c2fe138bebac224adddcdef8a7ce/blis-1.3.3-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:1e647341f958421a86b028a2efe16ce19c67dba2a05f79e8f7e80b1ff45328aa" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4a/78/750d12da388f714958eb2f2fd177652323bbe7ec528365c37129edd6eb84/blis-1.3.3-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:d563160f874abb78a57e346f07312c5323f7ad67b6370052b6b17087ef234a8e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e8/36/eac4199c5b200a5f3e93cad197da8d26d909f218eb444c4f552647c95240/blis-1.3.3-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:30b8a5b90cb6cb81d1ada9ae05aa55fb8e70d9a0ae9db40d2401bb9c1c8f14c4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/bf/51/472e7b36a6bedb5242a9757e7486f702c3619eff76e256735d0c8b1679c6/blis-1.3.3-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:e9f5c53b277f6ac5b3ca30bc12ebab7ea16c8f8c36b14428abb56924213dc127" },
{ url = "https://mirrors.aliyun.com/pypi/packages/84/da/d0dfb6d6e6321ae44df0321384c32c322bd07b15740d7422727a1a49fc5d/blis-1.3.3-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:6297e7616c158b305c9a8a4e47ca5fc9b0785194dd96c903b1a1591a7ca21ddf" },
{ url = "https://mirrors.aliyun.com/pypi/packages/20/c5/2b0b5e556fa0364ed671051ea078a6d6d7b979b1cfef78d64ad3ca5f0c7f/blis-1.3.3-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:3f966ca74f89f8a33e568b9a1d71992fc9a0d29a423e047f0a212643e21b5458" },
{ url = "https://mirrors.aliyun.com/pypi/packages/31/07/4cdc81a47bf862c0b06d91f1bc6782064e8b69ac9b5d4ff51d97e4ff03da/blis-1.3.3-cp313-cp313-win_amd64.whl", hash = "sha256:7a0fc4b237a3a453bdc3c7ab48d91439fcd2d013b665c46948d9eaf9c3e45a97" },
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "boto3"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.42.25"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "botocore" },
{ name = "jmespath" },
{ name = "s3transfer" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/29/30/755a6c4b27ad4effefa9e407f84c6f0a69f75a21c0090beb25022dfcfd3f/boto3-1.42.25.tar.gz", hash = "sha256:ccb5e757dd62698d25766cc54cf5c47bea43287efa59c93cf1df8c8fbc26eeda" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/14/79/012734f4e510b0a6beec2a3d5f437b3e8ef52174b1d38b1d5fdc542316d7/boto3-1.42.25-py3-none-any.whl", hash = "sha256:8128bde4f9d5ffce129c76d1a2efe220e3af967a2ad30bc305ba088bbc96343d" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "botocore"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.42.25"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "jmespath" },
{ name = "python-dateutil" },
{ name = "urllib3" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/2c/b5/8f961c65898deb5417c9e9e908ea6c4d2fe8bb52ff04e552f679c88ed2ce/botocore-1.42.25.tar.gz", hash = "sha256:7ae79d1f77d3771e83e4dd46bce43166a1ba85d58a49cffe4c4a721418616054" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/1e/b0/61e3e61d437c8c73f0821ce8a8e2594edfc1f423e354c38fa56396a4e4ca/botocore-1.42.25-py3-none-any.whl", hash = "sha256:470261966aab1d09a1cd4ba56810098834443602846559ba9504f6613dfa52dc" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: Add box connector (#11845) ### What problem does this PR solve? Feat: Add box connector ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-12 10:23:40 +08:00
[[package]]
name = "boxsdk"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "10.3.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Add box connector (#11845) ### What problem does this PR solve? Feat: Add box connector ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-12 10:23:40 +08:00
dependencies = [
{ name = "requests" },
{ name = "requests-toolbelt" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/df/25/d859cc617d832506e80327a277b0e0cc7d1114d66e966fdab8b218ffaf17/boxsdk-10.3.0.tar.gz", hash = "sha256:5b8ec0e2ed70160e16fe2fc1240d3896c88d50bd30796b021e95cfbe977b3444" }
Feat: Add box connector (#11845) ### What problem does this PR solve? Feat: Add box connector ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-12 10:23:40 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/05/af/fec6a530efdfc3d7739d821cdcb63de7c9979954fa21ef6d16d0b678c8ed/boxsdk-10.3.0-py3-none-any.whl", hash = "sha256:3f65792834315177765c096402e35f43400c4c99c9b6e82f9ac40c8de3da4767" },
Feat: Add box connector (#11845) ### What problem does this PR solve? Feat: Add box connector ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-12 10:23:40 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "brotli"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "1.2.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f7/16/c92ca344d646e71a43b8bb353f0a6490d7f6e06210f8554c8f874e454285/brotli-1.2.0.tar.gz", hash = "sha256:e310f77e41941c13340a95976fe66a8a95b01e783d430eeaf7a2f87e0a57dd0a" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/6c/d4/4ad5432ac98c73096159d9ce7ffeb82d151c2ac84adcc6168e476bb54674/brotli-1.2.0-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:9e5825ba2c9998375530504578fd4d5d1059d09621a02065d1b6bfc41a8e05ab" },
{ url = "https://mirrors.aliyun.com/pypi/packages/91/9f/9cc5bd03ee68a85dc4bc89114f7067c056a3c14b3d95f171918c088bf88d/brotli-1.2.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:0cf8c3b8ba93d496b2fae778039e2f5ecc7cff99df84df337ca31d8f2252896c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2e/b6/fe84227c56a865d16a6614e2c4722864b380cb14b13f3e6bef441e73a85a/brotli-1.2.0-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:c8565e3cdc1808b1a34714b553b262c5de5fbda202285782173ec137fd13709f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/55/de/de4ae0aaca06c790371cf6e7ee93a024f6b4bb0568727da8c3de112e726c/brotli-1.2.0-cp313-cp313-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:26e8d3ecb0ee458a9804f47f21b74845cc823fd1bb19f02272be70774f56e2a6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/5f/16/a1b22cbea436642e071adcaf8d4b350a2ad02f5e0ad0da879a1be16188a0/brotli-1.2.0-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:67a91c5187e1eec76a61625c77a6c8c785650f5b576ca732bd33ef58b0dff49c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/46/63/c968a97cbb3bdbf7f974ef5a6ab467a2879b82afbc5ffb65b8acbb744f95/brotli-1.2.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:4ecdb3b6dc36e6d6e14d3a1bdc6c1057c8cbf80db04031d566eb6080ce283a48" },
{ url = "https://mirrors.aliyun.com/pypi/packages/06/9d/102c67ea5c9fc171f423e8399e585dabea29b5bc79b05572891e70013cdd/brotli-1.2.0-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:3e1b35d56856f3ed326b140d3c6d9db91740f22e14b06e840fe4bb1923439a18" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9e/4a/9526d14fa6b87bc827ba1755a8440e214ff90de03095cacd78a64abe2b7d/brotli-1.2.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:54a50a9dad16b32136b2241ddea9e4df159b41247b2ce6aac0b3276a66a8f1e5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/5b/e8/3fe1ffed70cbef83c5236166acaed7bb9c766509b157854c80e2f766b38c/brotli-1.2.0-cp313-cp313-win32.whl", hash = "sha256:1b1d6a4efedd53671c793be6dd760fcf2107da3a52331ad9ea429edf0902f27a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ff/91/e739587be970a113b37b821eae8097aac5a48e5f0eca438c22e4c7dd8648/brotli-1.2.0-cp313-cp313-win_amd64.whl", hash = "sha256:b63daa43d82f0cdabf98dee215b375b4058cce72871fd07934f179885aad16e8" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
[[package]]
name = "browser-use"
version = "0.11.13"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "aiohttp" },
{ name = "anthropic" },
{ name = "anyio" },
{ name = "authlib" },
{ name = "browser-use-sdk" },
{ name = "bubus" },
{ name = "cdp-use" },
{ name = "click" },
{ name = "cloudpickle" },
{ name = "google-api-core" },
{ name = "google-api-python-client" },
{ name = "google-auth" },
{ name = "google-auth-oauthlib" },
{ name = "google-genai" },
{ name = "groq" },
{ name = "httpx" },
{ name = "inquirerpy" },
{ name = "markdownify" },
{ name = "mcp" },
{ name = "ollama" },
{ name = "openai" },
{ name = "pillow" },
{ name = "portalocker" },
{ name = "posthog" },
{ name = "psutil" },
{ name = "pydantic" },
{ name = "pyobjc", marker = "platform_system == 'darwin'" },
{ name = "pyotp" },
{ name = "pypdf" },
{ name = "python-docx" },
{ name = "python-dotenv" },
{ name = "reportlab" },
{ name = "requests" },
{ name = "rich" },
{ name = "screeninfo", marker = "platform_system != 'darwin'" },
{ name = "typing-extensions" },
{ name = "uuid7" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/9f/86/40464b112d01dfedf2433570a6537dea1656715bf8631d18a6eaa2dce28b/browser_use-0.11.13.tar.gz", hash = "sha256:c20d029f17c44add2047a72c836cb589b85e90a31a91cf3632a22a2de1928dfe" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/c6/ae/011c8a99708c82a2f8b75c5f24fb62541460fcb648050227db67d361bbe4/browser_use-0.11.13-py3-none-any.whl", hash = "sha256:f5232309213715e66e8f2079fb7097ac79a880728735968e4c7d41031ed15e83" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "browser-use-sdk"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "3.8.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "httpx" },
{ name = "pydantic" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/9b/c2/8194f96499faee7b070759b5773d0fa81916f9c670e7effbe306af419bc2/browser_use_sdk-3.8.1.tar.gz", hash = "sha256:c98a0f298d84e13f2f8eb07b7c7d7a6763dd42193844385954801e62f846fabb" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/e0/b9/ac19c900eb99e637e7d64a70daa367286a142ddc5a0645349f2f067699ed/browser_use_sdk-3.8.1-py3-none-any.whl", hash = "sha256:182280f3b64a549dec4c1e2cd8a59ebcd1965f5f87a2aea4f9e09de41c4ec9f7" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "bubus"
version = "1.5.6"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "aiofiles" },
{ name = "anyio" },
{ name = "portalocker" },
{ name = "pydantic" },
{ name = "typing-extensions" },
{ name = "uuid7" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/2d/85/aa72d1ffced7402fe41805519dab9935e9ce2bce18a10a55f2273ba8ba59/bubus-1.5.6.tar.gz", hash = "sha256:1a5456f0a576e86613a7bd66e819891b677778320b6e291094e339b0d9df2e0d" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/f5/54/23aae0681500a459fc4498b60754cb8ead8df964d8166e5915edb7e8136c/bubus-1.5.6-py3-none-any.whl", hash = "sha256:254ae37cd9299941f5e9d6afb11f8e3ce069f83e5b9476f88c6b2e32912f237d" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "cachelib"
version = "0.13.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/1d/69/0b5c1259e12fbcf5c2abe5934b5c0c1294ec0f845e2b4b2a51a91d79a4fb/cachelib-0.13.0.tar.gz", hash = "sha256:209d8996e3c57595bee274ff97116d1d73c4980b2fd9a34c7846cd07fd2e1a48" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/9b/42/960fc9896ddeb301716fdd554bab7941c35fb90a1dc7260b77df3366f87f/cachelib-0.13.0-py3-none-any.whl", hash = "sha256:8c8019e53b6302967d4e8329a504acf75e7bc46130291d30188a6e4e58162516" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "cachetools"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "6.2.4"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/bc/1d/ede8680603f6016887c062a2cf4fc8fdba905866a3ab8831aa8aa651320c/cachetools-6.2.4.tar.gz", hash = "sha256:82c5c05585e70b6ba2d3ae09ea60b79548872185d2f24ae1f2709d37299fd607" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/2c/fc/1d7b80d0eb7b714984ce40efc78859c022cd930e402f599d8ca9e39c78a4/cachetools-6.2.4-py3-none-any.whl", hash = "sha256:69a7a52634fed8b8bf6e24a050fb60bff1c9bd8f6d24572b99c32d4e71e62a51" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: add forgot password reset, solve #8547 (#10586) ### What problem does this PR solve? Feat: add forgot password reset, solve #8547 ### Type of change - [X] New Feature (non-breaking change which adds functionality)
2025-10-16 15:07:49 +08:00
[[package]]
name = "captcha"
version = "0.7.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: add forgot password reset, solve #8547 (#10586) ### What problem does this PR solve? Feat: add forgot password reset, solve #8547 ### Type of change - [X] New Feature (non-breaking change which adds functionality)
2025-10-16 15:07:49 +08:00
dependencies = [
{ name = "pillow" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/b4/65/8e186bb798f33ba390eab897c995b0fcee92bc030e0f40cb8ea01f34dd07/captcha-0.7.1.tar.gz", hash = "sha256:a1b462bcc633a64d8db5efa7754548a877c698d98f87716c620a707364cabd6b" }
Feat: add forgot password reset, solve #8547 (#10586) ### What problem does this PR solve? Feat: add forgot password reset, solve #8547 ### Type of change - [X] New Feature (non-breaking change which adds functionality)
2025-10-16 15:07:49 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/da/ff/3f0982ecd37c2d6a7266c22e7ea2e47d0773fe449984184c5316459d2776/captcha-0.7.1-py3-none-any.whl", hash = "sha256:8b73b5aba841ad1e5bdb856205bf5f09560b728ee890eb9dae42901219c8c599" },
Add Moodle data source integration (#11325) ### What problem does this PR solve? This PR adds a native Moodle connector to sync content (courses, resources, forums, assignments, pages, books) into RAGFlow. ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2025-11-21 12:58:49 +01:00
]
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
[[package]]
name = "catalogue"
version = "2.0.10"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/38/b4/244d58127e1cdf04cf2dc7d9566f0d24ef01d5ce21811bab088ecc62b5ea/catalogue-2.0.10.tar.gz", hash = "sha256:4f56daa940913d3f09d589c191c74e5a6d51762b3a9e37dd53b7437afd6cda15" }
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/9e/96/d32b941a501ab566a16358d68b6eb4e4acc373fab3c3c4d7d9e649f7b4bb/catalogue-2.0.10-py3-none-any.whl", hash = "sha256:58c2de0020aa90f4a2da7dfad161bf7b3b054c86a5f09fcedc0b2b740c109a9f" },
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
]
Add Moodle data source integration (#11325) ### What problem does this PR solve? This PR adds a native Moodle connector to sync content (courses, resources, forums, assignments, pages, books) into RAGFlow. ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2025-11-21 12:58:49 +01:00
[[package]]
name = "cattrs"
version = "22.2.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Add Moodle data source integration (#11325) ### What problem does this PR solve? This PR adds a native Moodle connector to sync content (courses, resources, forums, assignments, pages, books) into RAGFlow. ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2025-11-21 12:58:49 +01:00
dependencies = [
{ name = "attrs" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/fc/da/ff3239eb4241cbc6f8b69f53d4ca27a178d51f9e5a954f1a3588c8227dc5/cattrs-22.2.0.tar.gz", hash = "sha256:f0eed5642399423cf656e7b66ce92cdc5b963ecafd041d1b24d136fdde7acf6d" }
Add Moodle data source integration (#11325) ### What problem does this PR solve? This PR adds a native Moodle connector to sync content (courses, resources, forums, assignments, pages, books) into RAGFlow. ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2025-11-21 12:58:49 +01:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/43/3b/1d34fc4449174dfd2bc5ad7047a23edb6558b2e4b5a41b25a8ad6655c6c7/cattrs-22.2.0-py3-none-any.whl", hash = "sha256:bc12b1f0d000b9f9bee83335887d532a1d3e99a833d1bf0882151c97d3e68c21" },
Feat: add forgot password reset, solve #8547 (#10586) ### What problem does this PR solve? Feat: add forgot password reset, solve #8547 ### Type of change - [X] New Feature (non-breaking change which adds functionality)
2025-10-16 15:07:49 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "cbor"
version = "1.0.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/9b/99/01c6a987c920500189eb74a291bd3a388e6c7cf85736bb6b066d9833315e/cbor-1.0.0.tar.gz", hash = "sha256:13225a262ddf5615cbd9fd55a76a0d53069d18b07d2e9f19c39e6acb8609bbb6" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "cbor2"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "5.8.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/d9/8e/8b4fdde28e42ffcd741a37f4ffa9fb59cd4fe01625b544dfcfd9ccb54f01/cbor2-5.8.0.tar.gz", hash = "sha256:b19c35fcae9688ac01ef75bad5db27300c2537eb4ee00ed07e05d8456a0d4931" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/a6/0d/5a3f20bafaefeb2c1903d961416f051c0950f0d09e7297a3aa6941596b29/cbor2-5.8.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:6d8d104480845e2f28c6165b4c961bbe58d08cb5638f368375cfcae051c28015" },
{ url = "https://mirrors.aliyun.com/pypi/packages/57/66/177a3f089e69db69c987453ab4934086408c3338551e4984734597be9f80/cbor2-5.8.0-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:43efee947e5ab67d406d6e0dc61b5dee9d2f5e89ae176f90677a3741a20ca2e7" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b7/8e/9e17b8e4ed80a2ce97e2dfa5915c169dbb31599409ddb830f514b57f96cc/cbor2-5.8.0-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:be7ae582f50be539e09c134966d0fd63723fc4789b8dff1f6c2e3f24ae3eaf32" },
{ url = "https://mirrors.aliyun.com/pypi/packages/cc/33/9f92e107d78f88ac22723ac15d0259d220ba98c1d855e51796317f4c4114/cbor2-5.8.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:50f5c709561a71ea7970b4cd2bf9eda4eccacc0aac212577080fdfe64183e7f5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2f/3f/46b80050a4a35ce5cf7903693864a9fdea7213567dc8faa6e25cb375c182/cbor2-5.8.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:a6790ecc73aa93e76d2d9076fc42bf91a9e69f2295e5fa702e776dbe986465bd" },
{ url = "https://mirrors.aliyun.com/pypi/packages/eb/d2/d41f8c04c783a4d204e364be2d38043d4f732a3bed6f4c732e321cf34c7b/cbor2-5.8.0-cp313-cp313-win_amd64.whl", hash = "sha256:c114af8099fa65a19a514db87ce7a06e942d8fea2730afd49be39f8e16e7f5e0" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1b/8c/0397a82f6e67665009951453c83058e4c77ba54b9a9017ede56d6870306c/cbor2-5.8.0-cp313-cp313-win_arm64.whl", hash = "sha256:ab3ba00494ad8669a459b12a558448d309c271fa4f89b116ad496ee35db38fea" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d6/4f/101071f880b4da05771128c0b89f41e334cff044dee05fb013c8f4be661c/cbor2-5.8.0-py3-none-any.whl", hash = "sha256:3727d80f539567b03a7aa11890e57798c67092c38df9e6c23abb059e0f65069c" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
[[package]]
name = "cdp-use"
version = "1.4.5"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "httpx" },
{ name = "typing-extensions" },
{ name = "websockets" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f7/7a/c549417e8c5e4dface6d5d828cd7dc72502dcea33a99f5324abf5a853ce9/cdp_use-1.4.5.tar.gz", hash = "sha256:0da3a32df46336a03ff5a22bc6bc442cd7d2f2d50a118fd4856f29d37f6d26a0" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/56/12/386d8c6bf0448c43674e24d6194c3b57d62e5361e90bca3d58108819ad32/cdp_use-1.4.5-py3-none-any.whl", hash = "sha256:8f8e2435e3a20e4009d2974144192cf3c132f6c2971338e156198814d9b91ecb" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "certifi"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "2026.1.4"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/e0/2d/a891ca51311197f6ad14a7ef42e2399f36cf2f9bd44752b3dc4eab60fdc5/certifi-2026.1.4.tar.gz", hash = "sha256:ac726dd470482006e014ad384921ed6438c457018f4b3d204aea4281258b2120" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/e6/ad/3cc14f097111b4de0040c83a525973216457bbeeb63739ef1ed275c1c021/certifi-2026.1.4-py3-none-any.whl", hash = "sha256:9943707519e4add1115f44c2bc244f782c0249876bf51b6599fee1ffbedd685c" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "cffi"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "2.0.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
dependencies = [
{ name = "pycparser", marker = "implementation_name != 'PyPy'" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/eb/56/b1ba7935a17738ae8453301356628e8147c79dbb825bcbc73dc7401f9846/cffi-2.0.0.tar.gz", hash = "sha256:44d1b5909021139fe36001ae048dbdde8214afa20200eda0f64c068cac5d5529" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/4b/8d/a0a47a0c9e413a658623d014e91e74a50cdd2c423f7ccfd44086ef767f90/cffi-2.0.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:00bdf7acc5f795150faa6957054fbbca2439db2f775ce831222b66f192f03beb" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4a/d2/a6c0296814556c68ee32009d9c2ad4f85f2707cdecfd7727951ec228005d/cffi-2.0.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:45d5e886156860dc35862657e1494b9bae8dfa63bf56796f2fb56e1679fc0bca" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b0/1e/d22cc63332bd59b06481ceaac49d6c507598642e2230f201649058a7e704/cffi-2.0.0-cp313-cp313-manylinux1_i686.manylinux2014_i686.manylinux_2_17_i686.manylinux_2_5_i686.whl", hash = "sha256:07b271772c100085dd28b74fa0cd81c8fb1a3ba18b21e03d7c27f3436a10606b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a9/f5/a2c23eb03b61a0b8747f211eb716446c826ad66818ddc7810cc2cc19b3f2/cffi-2.0.0-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:d48a880098c96020b02d5a1f7d9251308510ce8858940e6fa99ece33f610838b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f2/7f/e6647792fc5850d634695bc0e6ab4111ae88e89981d35ac269956605feba/cffi-2.0.0-cp313-cp313-manylinux2014_ppc64le.manylinux_2_17_ppc64le.whl", hash = "sha256:f93fd8e5c8c0a4aa1f424d6173f14a892044054871c771f8566e4008eaa359d2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/cb/1e/a5a1bd6f1fb30f22573f76533de12a00bf274abcdc55c8edab639078abb6/cffi-2.0.0-cp313-cp313-manylinux2014_s390x.manylinux_2_17_s390x.whl", hash = "sha256:dd4f05f54a52fb558f1ba9f528228066954fee3ebe629fc1660d874d040ae5a3" },
{ url = "https://mirrors.aliyun.com/pypi/packages/98/df/0a1755e750013a2081e863e7cd37e0cdd02664372c754e5560099eb7aa44/cffi-2.0.0-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:c8d3b5532fc71b7a77c09192b4a5a200ea992702734a2e9279a37f2478236f26" },
{ url = "https://mirrors.aliyun.com/pypi/packages/50/e1/a969e687fcf9ea58e6e2a928ad5e2dd88cc12f6f0ab477e9971f2309b57c/cffi-2.0.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:d9b29c1f0ae438d5ee9acb31cadee00a58c46cc9c0b2f9038c6b0b3470877a8c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/36/54/0362578dd2c9e557a28ac77698ed67323ed5b9775ca9d3fe73fe191bb5d8/cffi-2.0.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:6d50360be4546678fc1b79ffe7a66265e28667840010348dd69a314145807a1b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/eb/6d/bf9bda840d5f1dfdbf0feca87fbdb64a918a69bca42cfa0ba7b137c48cb8/cffi-2.0.0-cp313-cp313-win32.whl", hash = "sha256:74a03b9698e198d47562765773b4a8309919089150a0bb17d829ad7b44b60d27" },
{ url = "https://mirrors.aliyun.com/pypi/packages/37/18/6519e1ee6f5a1e579e04b9ddb6f1676c17368a7aba48299c3759bbc3c8b3/cffi-2.0.0-cp313-cp313-win_amd64.whl", hash = "sha256:19f705ada2530c1167abacb171925dd886168931e0a7b78f5bffcae5c6b5be75" },
{ url = "https://mirrors.aliyun.com/pypi/packages/cb/0e/02ceeec9a7d6ee63bb596121c2c8e9b3a9e150936f4fbef6ca1943e6137c/cffi-2.0.0-cp313-cp313-win_arm64.whl", hash = "sha256:256f80b80ca3853f90c21b23ee78cd008713787b1b1e93eae9f3d6a7134abd91" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "chardet"
version = "5.2.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f3/0d/f7b6ab21ec75897ed80c17d79b15951a719226b9fababf1e40ea74d69079/chardet-5.2.0.tar.gz", hash = "sha256:1b3b6ff479a8c414bc3fa2c0852995695c4a026dcd6d0633b2dd092ca39c1cf7" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/38/6f/f5fbc992a329ee4e0f288c1fe0e2ad9485ed064cac731ed2fe47dcc38cbf/chardet-5.2.0-py3-none-any.whl", hash = "sha256:e1cf59446890a00105fe7b7912492ea04b6e6f06d4b742b2c788469e34c82970" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "charset-normalizer"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "3.4.4"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/13/69/33ddede1939fdd074bce5434295f38fae7136463422fe4fd3e0e89b98062/charset_normalizer-3.4.4.tar.gz", hash = "sha256:94537985111c35f28720e43603b8e7b43a6ecfb2ce1d3058bbe955b73404e21a" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/97/45/4b3a1239bbacd321068ea6e7ac28875b03ab8bc0aa0966452db17cd36714/charset_normalizer-3.4.4-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:e1f185f86a6f3403aa2420e815904c67b2f9ebc443f045edd0de921108345794" },
{ url = "https://mirrors.aliyun.com/pypi/packages/7d/62/73a6d7450829655a35bb88a88fca7d736f9882a27eacdca2c6d505b57e2e/charset_normalizer-3.4.4-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:6b39f987ae8ccdf0d2642338faf2abb1862340facc796048b604ef14919e55ed" },
{ url = "https://mirrors.aliyun.com/pypi/packages/89/c5/adb8c8b3d6625bef6d88b251bbb0d95f8205831b987631ab0c8bb5d937c2/charset_normalizer-3.4.4-cp313-cp313-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:3162d5d8ce1bb98dd51af660f2121c55d0fa541b46dff7bb9b9f86ea1d87de72" },
{ url = "https://mirrors.aliyun.com/pypi/packages/91/ed/9706e4070682d1cc219050b6048bfd293ccf67b3d4f5a4f39207453d4b99/charset_normalizer-3.4.4-cp313-cp313-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:81d5eb2a312700f4ecaa977a8235b634ce853200e828fbadf3a9c50bab278328" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d5/0d/031f0d95e4972901a2f6f09ef055751805ff541511dc1252ba3ca1f80cf5/charset_normalizer-3.4.4-cp313-cp313-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:5bd2293095d766545ec1a8f612559f6b40abc0eb18bb2f5d1171872d34036ede" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f5/83/6ab5883f57c9c801ce5e5677242328aa45592be8a00644310a008d04f922/charset_normalizer-3.4.4-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:a8a8b89589086a25749f471e6a900d3f662d1d3b6e2e59dcecf787b1cc3a1894" },
{ url = "https://mirrors.aliyun.com/pypi/packages/75/1e/5ff781ddf5260e387d6419959ee89ef13878229732732ee73cdae01800f2/charset_normalizer-3.4.4-cp313-cp313-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:bc7637e2f80d8530ee4a78e878bce464f70087ce73cf7c1caf142416923b98f1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d7/57/71be810965493d3510a6ca79b90c19e48696fb1ff964da319334b12677f0/charset_normalizer-3.4.4-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:f8bf04158c6b607d747e93949aa60618b61312fe647a6369f88ce2ff16043490" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e5/d5/c3d057a78c181d007014feb7e9f2e65905a6c4ef182c0ddf0de2924edd65/charset_normalizer-3.4.4-cp313-cp313-musllinux_1_2_armv7l.whl", hash = "sha256:554af85e960429cf30784dd47447d5125aaa3b99a6f0683589dbd27e2f45da44" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e6/8c/d0406294828d4976f275ffbe66f00266c4b3136b7506941d87c00cab5272/charset_normalizer-3.4.4-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:74018750915ee7ad843a774364e13a3db91682f26142baddf775342c3f5b1133" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d7/24/e2aa1f18c8f15c4c0e932d9287b8609dd30ad56dbe41d926bd846e22fb8d/charset_normalizer-3.4.4-cp313-cp313-musllinux_1_2_riscv64.whl", hash = "sha256:c0463276121fdee9c49b98908b3a89c39be45d86d1dbaa22957e38f6321d4ce3" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e4/5b/1e6160c7739aad1e2df054300cc618b06bf784a7a164b0f238360721ab86/charset_normalizer-3.4.4-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:362d61fd13843997c1c446760ef36f240cf81d3ebf74ac62652aebaf7838561e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/7a/10/f882167cd207fbdd743e55534d5d9620e095089d176d55cb22d5322f2afd/charset_normalizer-3.4.4-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:9a26f18905b8dd5d685d6d07b0cdf98a79f3c7a918906af7cc143ea2e164c8bc" },
{ url = "https://mirrors.aliyun.com/pypi/packages/89/66/c7a9e1b7429be72123441bfdbaf2bc13faab3f90b933f664db506dea5915/charset_normalizer-3.4.4-cp313-cp313-win32.whl", hash = "sha256:9b35f4c90079ff2e2edc5b26c0c77925e5d2d255c42c74fdb70fb49b172726ac" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c4/26/b9924fa27db384bdcd97ab83b4f0a8058d96ad9626ead570674d5e737d90/charset_normalizer-3.4.4-cp313-cp313-win_amd64.whl", hash = "sha256:b435cba5f4f750aa6c0a0d92c541fb79f69a387c91e61f1795227e4ed9cece14" },
{ url = "https://mirrors.aliyun.com/pypi/packages/af/8f/3ed4bfa0c0c72a7ca17f0380cd9e4dd842b09f664e780c13cff1dcf2ef1b/charset_normalizer-3.4.4-cp313-cp313-win_arm64.whl", hash = "sha256:542d2cee80be6f80247095cc36c418f7bddd14f4a6de45af91dfad36d817bba2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0a/4c/925909008ed5a988ccbb72dcc897407e5d6d3bd72410d69e051fc0c14647/charset_normalizer-3.4.4-py3-none-any.whl", hash = "sha256:7a32c560861a02ff789ad905a2fe94e3f840803362c84fecf1851cb4cf3dc37f" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "click"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "8.3.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "colorama", marker = "sys_platform == 'win32'" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/3d/fa/656b739db8587d7b5dfa22e22ed02566950fbfbcdc20311993483657a5c0/click-8.3.1.tar.gz", hash = "sha256:12ff4785d337a1bb490bb7e9c2b1ee5da3112e94a8622f26a6c77f5d2fc6842a" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/98/78/01c019cdb5d6498122777c1a43056ebb3ebfeef2076d9d026bfe15583b2b/click-8.3.1-py3-none-any.whl", hash = "sha256:981153a64e25f12d547d3426c367a4857371575ee7ad18df2a6183ab0545b2a6" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
[[package]]
name = "click-log"
version = "0.4.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
dependencies = [
{ name = "click" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/32/32/228be4f971e4bd556c33d52a22682bfe318ffe57a1ddb7a546f347a90260/click-log-0.4.0.tar.gz", hash = "sha256:3970f8570ac54491237bcdb3d8ab5e3eef6c057df29f8c3d1151a51a9c23b975" }
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ae/5a/4f025bc751087833686892e17e7564828e409c43b632878afeae554870cd/click_log-0.4.0-py2.py3-none-any.whl", hash = "sha256:a43e394b528d52112af599f2fc9e4b7cf3c15f94e53581f74fa6867e68c91756" },
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
]
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
[[package]]
fix: upgrade crawl4ai to 0.8.0 (CVE-2026-26217) (#15415) ## Summary Upgrade crawl4ai from 0.7.6 to 0.8.0 to fix CVE-2026-26217. ## Vulnerability | Field | Value | |-------|-------| | **ID** | CVE-2026-26217 | | **Severity** | CRITICAL | | **Scanner** | trivy | | **Rule** | `CVE-2026-26217` | | **File** | `uv.lock` | | **Assessment** | Likely exploitable | **Description**: Crawl4AI Has Local File Inclusion in Docker API via file:// URLs ## Evidence **Scanner confirmation**: trivy rule `CVE-2026-26217` flagged this pattern. **Production code**: This file is in the production codebase, not test-only code. ## Threat Model Context This is a web service - vulnerabilities in request handlers are directly exploitable by remote attackers. ## Changes - `pyproject.toml` - `uv.lock` ## Verification - [x] Build passes - [x] Scanner re-scan confirms fix - [x] LLM code review passed --- *This change addresses a pattern flagged by static analysis. The code path handles user-influenced input and the fix reduces the attack surface against both manual and automated exploitation.* --- *Automated security fix by [OrbisAI Security](https://orbisappsec.com)*
2026-05-29 19:08:41 +05:30
name = "cloudpathlib"
version = "0.24.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/06/19/58bc6b5d7d0f81c7209b05445af477e147c486552f96665a5912211839b9/cloudpathlib-0.24.0.tar.gz", hash = "sha256:c521a984e77b47e656fe78e20a7e3e260e0ab45fc69e33ac01094227c979e34a" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/c2/5b/ba933f896d9b0b07608d575a8501e2b4e32166b60d84c430a4a7285ebe64/cloudpathlib-0.24.0-py3-none-any.whl", hash = "sha256:b1c51e2d2ec7dc4fed6538991f4aea849d6cf11a7e6b9069f86e461aa1f9b5b4" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
[[package]]
fix: upgrade crawl4ai to 0.8.0 (CVE-2026-26217) (#15415) ## Summary Upgrade crawl4ai from 0.7.6 to 0.8.0 to fix CVE-2026-26217. ## Vulnerability | Field | Value | |-------|-------| | **ID** | CVE-2026-26217 | | **Severity** | CRITICAL | | **Scanner** | trivy | | **Rule** | `CVE-2026-26217` | | **File** | `uv.lock` | | **Assessment** | Likely exploitable | **Description**: Crawl4AI Has Local File Inclusion in Docker API via file:// URLs ## Evidence **Scanner confirmation**: trivy rule `CVE-2026-26217` flagged this pattern. **Production code**: This file is in the production codebase, not test-only code. ## Threat Model Context This is a web service - vulnerabilities in request handlers are directly exploitable by remote attackers. ## Changes - `pyproject.toml` - `uv.lock` ## Verification - [x] Build passes - [x] Scanner re-scan confirms fix - [x] LLM code review passed --- *This change addresses a pattern flagged by static analysis. The code path handles user-influenced input and the fix reduces the attack surface against both manual and automated exploitation.* --- *Automated security fix by [OrbisAI Security](https://orbisappsec.com)*
2026-05-29 19:08:41 +05:30
name = "cloudpickle"
version = "3.1.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/27/fb/576f067976d320f5f0114a8d9fa1215425441bb35627b1993e5afd8111e5/cloudpickle-3.1.2.tar.gz", hash = "sha256:7fda9eb655c9c230dab534f1983763de5835249750e85fbcef43aaa30a9a2414" }
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/88/39/799be3f2f0f38cc727ee3b4f1445fe6d5e4133064ec2e4115069418a5bb6/cloudpickle-3.1.2-py3-none-any.whl", hash = "sha256:9acb47f6afd73f60dc1df93bb801b472f05ff42fa6c84167d25cb206be1fbf4a" },
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "cn2an"
version = "0.5.22"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "proces" },
{ name = "setuptools" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a1/5c/f565259e568316e5fde4dba292e2ca9cff0619657e4ec9f254f415543f59/cn2an-0.5.22.tar.gz", hash = "sha256:27ae5b56441d7329ed2ececffa026bfa8fc353dcf1fb0d9146b303b9cce3ac37" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/1c/3d/3e04a822b8615904269f7126d8b019ae5c3b5c3c78397ec8bab056b02099/cn2an-0.5.22-py3-none-any.whl", hash = "sha256:cba4c8f305b43da01f50696047cca3116c727424ac62338da6a3426e01454f3e" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
[[package]]
name = "cobble"
version = "0.1.4"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/54/7a/a507c709be2c96e1bb6102eb7b7f4026c5e5e223ef7d745a17d239e9d844/cobble-0.1.4.tar.gz", hash = "sha256:de38be1539992c8a06e569630717c485a5f91be2192c461ea2b220607dfa78aa" }
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/d5/e1/3714a2f371985215c219c2a70953d38e3eed81ef165aed061d21de0e998b/cobble-0.1.4-py3-none-any.whl", hash = "sha256:36c91b1655e599fd428e2b95fdd5f0da1ca2e9f1abb0bc871dec21a0e78a2b44" },
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
]
Test: Add code coverage reporting to CI (#12874) ### What problem does this PR solve? Add code coverage reporting to CI ### Type of change - [x] Test (please describe): coverage report --------- Co-authored-by: Liu An <asiro@qq.com>
2026-01-30 14:49:16 +08:00
[[package]]
name = "codecov"
version = "2.1.13"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Test: Add code coverage reporting to CI (#12874) ### What problem does this PR solve? Add code coverage reporting to CI ### Type of change - [x] Test (please describe): coverage report --------- Co-authored-by: Liu An <asiro@qq.com>
2026-01-30 14:49:16 +08:00
dependencies = [
{ name = "coverage" },
{ name = "requests" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/2c/bb/594b26d2c85616be6195a64289c578662678afa4910cef2d3ce8417cf73e/codecov-2.1.13.tar.gz", hash = "sha256:2362b685633caeaf45b9951a9b76ce359cd3581dd515b430c6c3f5dfb4d92a8c" }
Test: Add code coverage reporting to CI (#12874) ### What problem does this PR solve? Add code coverage reporting to CI ### Type of change - [x] Test (please describe): coverage report --------- Co-authored-by: Liu An <asiro@qq.com>
2026-01-30 14:49:16 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/af/02/18785edcdf6266cdd6c6dc7635f1cbeefd9a5b4c3bb8aff8bd681e9dd095/codecov-2.1.13-py2.py3-none-any.whl", hash = "sha256:c2ca5e51bba9ebb43644c43d0690148a55086f7f5e6fd36170858fa4206744d5" },
Test: Add code coverage reporting to CI (#12874) ### What problem does this PR solve? Add code coverage reporting to CI ### Type of change - [x] Test (please describe): coverage report --------- Co-authored-by: Liu An <asiro@qq.com>
2026-01-30 14:49:16 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "cohere"
version = "5.6.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "boto3" },
{ name = "fastavro" },
{ name = "httpx" },
{ name = "httpx-sse" },
{ name = "parameterized" },
{ name = "pydantic" },
{ name = "requests" },
{ name = "tokenizers" },
{ name = "types-requests" },
{ name = "typing-extensions" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/2d/e0/d821fe7b3c0b30893b89a22f4e58d431211156499ca00805568b90aafcf6/cohere-5.6.2.tar.gz", hash = "sha256:6bb901afdfb02f62ad8ed2d82f12d8ea87a6869710f5f880cb89190c4e994805" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/b2/a7/0572d6ab1d947bd11aa8fc40fe908635fabc5abf254175943c2228c9d108/cohere-5.6.2-py3-none-any.whl", hash = "sha256:cfecf1343bcaa4091266c5a231fbcb3ccbd80cad05ea093ef80024a117aa3a2f" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "colorama"
version = "0.4.6"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/d8/53/6f443c9a4a8358a93a6792e2acffb9d9d5cb0a5cfd8802644b7b1c9a02e4/colorama-0.4.6.tar.gz", hash = "sha256:08695f5cb7ed6e0531a20572697297273c47b8cae5a63ffc6d6ed5c201be6e44" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/d1/d6/3965ed04c63042e047cb6a3e6ed1a63a35087b6a609aa3a15ed8ac56c221/colorama-0.4.6-py2.py3-none-any.whl", hash = "sha256:4f1d9991f5acc0ca119f9d443620b77f9d6b33703e51011c16baf57afb285fc6" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
[[package]]
name = "colorclass"
version = "2.2.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/d7/1a/31ff00a33569a3b59d65bbdc445c73e12f92ad28195b7ace299f68b9af70/colorclass-2.2.2.tar.gz", hash = "sha256:6d4fe287766166a98ca7bc6f6312daf04a0481b1eda43e7173484051c0ab4366" }
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/30/b6/daf3e2976932da4ed3579cff7a30a53d22ea9323ee4f0d8e43be60454897/colorclass-2.2.2-py2.py3-none-any.whl", hash = "sha256:6f10c273a0ef7a1150b1120b6095cbdd68e5cf36dfd5d0fc957a2500bbf99a55" },
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "coloredlogs"
version = "15.0.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
Fix onnxruntime-gpu marks (#4643) ### What problem does this PR solve? Fix onnxruntime-gpu marks ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-01-26 09:37:59 +08:00
{ name = "humanfriendly" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/cc/c7/eed8f27100517e8c0e6b923d5f0845d0cb99763da6fdee00478f91db7325/coloredlogs-15.0.1.tar.gz", hash = "sha256:7c991aa71a4577af2f82600d8f8f3a89f936baeaf9b50a9c197da014e5bf16b0" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/a7/06/3d6badcf13db419e25b07041d9c7b4a2c331d3f4e7134445ec5df57714cd/coloredlogs-15.0.1-py2.py3-none-any.whl", hash = "sha256:612ee75c546f53e92e70049c9dbfcc18c935a2b9a53b66085ce9ef6a6e5c0934" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
[[package]]
name = "compressed-rtf"
version = "1.0.7"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/b7/0c/929a4e8ef9d7143f54d77dadb5f370cc7b98534b1bd6e1124d0abe8efb24/compressed_rtf-1.0.7.tar.gz", hash = "sha256:7c30859334839f3cdc7d10796af5b434bb326b9df7cb5a65e95a8eacb2951b0e" }
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/07/1d/62f5bf92e12335eb63517f42671ed78512d48bbc69e02a942dd7b90f03f0/compressed_rtf-1.0.7-py3-none-any.whl", hash = "sha256:b7904921d78c67a0a4b7fff9fb361a00ae2b447b6edca010ce321cd98fa0fcc0" },
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
]
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
[[package]]
name = "confection"
version = "1.3.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ca/65/efd0fe8a936fc8ca2978cb7b82581fb20d901c6039e746a808f746b7647b/confection-1.3.3.tar.gz", hash = "sha256:f0f6810d567ff73993fe74d218ca5e1ffb6a44fb03f391257fc5d033546cbfaa" }
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/8d/e4/d66708bdf0d92fb4d49b22cdff4b10cec38aca5dcd7e81d909bb55c65cd7/confection-1.3.3-py3-none-any.whl", hash = "sha256:b9fef9ee84b237ef4611ec3eb5797b70e13063e6310ad9f15536373f5e313c82" },
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
]
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
[[package]]
name = "contourpy"
version = "1.3.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
dependencies = [
Bump infinity to v0.6.11. Requires python>=3.11 (#11814) ### What problem does this PR solve? Bump infinity to v0.6.11. Requires python>=3.11 ### Type of change - [x] Refactoring
2025-12-09 16:23:37 +08:00
{ name = "numpy" },
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/58/01/1253e6698a07380cd31a736d248a3f2a50a7c88779a1813da27503cadc2a/contourpy-1.3.3.tar.gz", hash = "sha256:083e12155b210502d0bca491432bb04d56dc3432f95a979b429f2848c3dbe880" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/68/35/0167aad910bbdb9599272bd96d01a9ec6852f36b9455cf2ca67bd4cc2d23/contourpy-1.3.3-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:177fb367556747a686509d6fef71d221a4b198a3905fe824430e5ea0fda54eb5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/96/e4/7adcd9c8362745b2210728f209bfbcf7d91ba868a2c5f40d8b58f54c509b/contourpy-1.3.3-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:d002b6f00d73d69333dac9d0b8d5e84d9724ff9ef044fd63c5986e62b7c9e1b1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/73/23/90e31ceeed1de63058a02cb04b12f2de4b40e3bef5e082a7c18d9c8ae281/contourpy-1.3.3-cp313-cp313-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:348ac1f5d4f1d66d3322420f01d42e43122f43616e0f194fc1c9f5d830c5b286" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ed/93/b43d8acbe67392e659e1d984700e79eb67e2acb2bd7f62012b583a7f1b55/contourpy-1.3.3-cp313-cp313-manylinux_2_26_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:655456777ff65c2c548b7c454af9c6f33f16c8884f11083244b5819cc214f1b5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/46/3b/bec82a3ea06f66711520f75a40c8fc0b113b2a75edb36aa633eb11c4f50f/contourpy-1.3.3-cp313-cp313-manylinux_2_26_s390x.manylinux_2_28_s390x.whl", hash = "sha256:644a6853d15b2512d67881586bd03f462c7ab755db95f16f14d7e238f2852c67" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4b/32/e0f13a1c5b0f8572d0ec6ae2f6c677b7991fafd95da523159c19eff0696a/contourpy-1.3.3-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:4debd64f124ca62069f313a9cb86656ff087786016d76927ae2cf37846b006c9" },
{ url = "https://mirrors.aliyun.com/pypi/packages/33/71/e2a7945b7de4e58af42d708a219f3b2f4cff7386e6b6ab0a0fa0033c49a9/contourpy-1.3.3-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:a15459b0f4615b00bbd1e91f1b9e19b7e63aea7483d03d804186f278c0af2659" },
{ url = "https://mirrors.aliyun.com/pypi/packages/12/fc/4e87ac754220ccc0e807284f88e943d6d43b43843614f0a8afa469801db0/contourpy-1.3.3-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:ca0fdcd73925568ca027e0b17ab07aad764be4706d0a925b89227e447d9737b7" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a6/2e/adc197a37443f934594112222ac1aa7dc9a98faf9c3842884df9a9d8751d/contourpy-1.3.3-cp313-cp313-win32.whl", hash = "sha256:b20c7c9a3bf701366556e1b1984ed2d0cedf999903c51311417cf5f591d8c78d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/18/0b/0098c214843213759692cc638fce7de5c289200a830e5035d1791d7a2338/contourpy-1.3.3-cp313-cp313-win_amd64.whl", hash = "sha256:1cadd8b8969f060ba45ed7c1b714fe69185812ab43bd6b86a9123fe8f99c3263" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8a/9a/2f6024a0c5995243cd63afdeb3651c984f0d2bc727fd98066d40e141ad73/contourpy-1.3.3-cp313-cp313-win_arm64.whl", hash = "sha256:fd914713266421b7536de2bfa8181aa8c699432b6763a0ea64195ebe28bff6a9" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c0/b3/f8a1a86bd3298513f500e5b1f5fd92b69896449f6cab6a146a5d52715479/contourpy-1.3.3-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:88df9880d507169449d434c293467418b9f6cbe82edd19284aa0409e7fdb933d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3f/11/4780db94ae62fc0c2053909b65dc3246bd7cecfc4f8a20d957ad43aa4ad8/contourpy-1.3.3-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:d06bb1f751ba5d417047db62bca3c8fde202b8c11fb50742ab3ab962c81e8216" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ae/15/e59f5f3ffdd6f3d4daa3e47114c53daabcb18574a26c21f03dc9e4e42ff0/contourpy-1.3.3-cp313-cp313t-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:e4e6b05a45525357e382909a4c1600444e2a45b4795163d3b22669285591c1ae" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0f/81/03b45cfad088e4770b1dcf72ea78d3802d04200009fb364d18a493857210/contourpy-1.3.3-cp313-cp313t-manylinux_2_26_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:ab3074b48c4e2cf1a960e6bbeb7f04566bf36b1861d5c9d4d8ac04b82e38ba20" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0c/ba/49923366492ffbdd4486e970d421b289a670ae8cf539c1ea9a09822b371a/contourpy-1.3.3-cp313-cp313t-manylinux_2_26_s390x.manylinux_2_28_s390x.whl", hash = "sha256:6c3d53c796f8647d6deb1abe867daeb66dcc8a97e8455efa729516b997b8ed99" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9f/52/5b00ea89525f8f143651f9f03a0df371d3cbd2fccd21ca9b768c7a6500c2/contourpy-1.3.3-cp313-cp313t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:50ed930df7289ff2a8d7afeb9603f8289e5704755c7e5c3bbd929c90c817164b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/32/1d/a209ec1a3a3452d490f6b14dd92e72280c99ae3d1e73da74f8277d4ee08f/contourpy-1.3.3-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:4feffb6537d64b84877da813a5c30f1422ea5739566abf0bd18065ac040e120a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/bc/9e/46f0e8ebdd884ca0e8877e46a3f4e633f6c9c8c4f3f6e72be3fe075994aa/contourpy-1.3.3-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:2b7e9480ffe2b0cd2e787e4df64270e3a0440d9db8dc823312e2c940c167df7e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b9/70/f308384a3ae9cd2209e0849f33c913f658d3326900d0ff5d378d6a1422d2/contourpy-1.3.3-cp313-cp313t-win32.whl", hash = "sha256:283edd842a01e3dcd435b1c5116798d661378d83d36d337b8dde1d16a5fc9ba3" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b2/dd/880f890a6663b84d9e34a6f88cded89d78f0091e0045a284427cb6b18521/contourpy-1.3.3-cp313-cp313t-win_amd64.whl", hash = "sha256:87acf5963fc2b34825e5b6b048f40e3635dd547f590b04d2ab317c2619ef7ae8" },
{ url = "https://mirrors.aliyun.com/pypi/packages/80/99/2adc7d8ffead633234817ef8e9a87115c8a11927a94478f6bb3d3f4d4f7d/contourpy-1.3.3-cp313-cp313t-win_arm64.whl", hash = "sha256:3c30273eb2a55024ff31ba7d052dde990d7d8e5450f4bbb6e913558b3d6c2301" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Add ut test into CI (#11753) ### What problem does this PR solve? As title ### Type of change - [x] Other (please describe): --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-05 11:40:16 +08:00
[[package]]
name = "coverage"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "7.13.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/23/f9/e92df5e07f3fc8d4c7f9a0f146ef75446bf870351cd37b788cf5897f8079/coverage-7.13.1.tar.gz", hash = "sha256:b7593fe7eb5feaa3fbb461ac79aac9f9fc0387a5ca8080b0c6fe2ca27b091afd" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/a3/a4/e98e689347a1ff1a7f67932ab535cef82eb5e78f32a9e4132e114bbb3a0a/coverage-7.13.1-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:cb237bfd0ef4d5eb6a19e29f9e528ac67ac3be932ea6b44fb6cc09b9f3ecff78" },
{ url = "https://mirrors.aliyun.com/pypi/packages/32/33/7cbfe2bdc6e2f03d6b240d23dc45fdaf3fd270aaf2d640be77b7f16989ab/coverage-7.13.1-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:1dcb645d7e34dcbcc96cd7c132b1fc55c39263ca62eb961c064eb3928997363b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/59/f6/efdabdb4929487baeb7cb2a9f7dac457d9356f6ad1b255be283d58b16316/coverage-7.13.1-cp313-cp313-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:3d42df8201e00384736f0df9be2ced39324c3907607d17d50d50116c989d84cd" },
{ url = "https://mirrors.aliyun.com/pypi/packages/12/da/91a52516e9d5aea87d32d1523f9cdcf7a35a3b298e6be05d6509ba3cfab2/coverage-7.13.1-cp313-cp313-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:fa3edde1aa8807de1d05934982416cb3ec46d1d4d91e280bcce7cca01c507992" },
{ url = "https://mirrors.aliyun.com/pypi/packages/75/38/f1ea837e3dc1231e086db1638947e00d264e7e8c41aa8ecacf6e1e0c05f4/coverage-7.13.1-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:9edd0e01a343766add6817bc448408858ba6b489039eaaa2018474e4001651a4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/7f/43/f4f16b881aaa34954ba446318dea6b9ed5405dd725dd8daac2358eda869a/coverage-7.13.1-cp313-cp313-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:985b7836931d033570b94c94713c6dba5f9d3ff26045f72c3e5dbc5fe3361e5a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/84/34/8cba7f00078bd468ea914134e0144263194ce849ec3baad187ffb6203d1c/coverage-7.13.1-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:ffed1e4980889765c84a5d1a566159e363b71d6b6fbaf0bebc9d3c30bc016766" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8c/a4/cffac66c7652d84ee4ac52d3ccb94c015687d3b513f9db04bfcac2ac800d/coverage-7.13.1-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:8842af7f175078456b8b17f1b73a0d16a65dcbdc653ecefeb00a56b3c8c298c4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f4/78/9a64d462263dde416f3c0067efade7b52b52796f489b1037a95b0dc389c9/coverage-7.13.1-cp313-cp313-musllinux_1_2_riscv64.whl", hash = "sha256:ccd7a6fca48ca9c131d9b0a2972a581e28b13416fc313fb98b6d24a03ce9a398" },
{ url = "https://mirrors.aliyun.com/pypi/packages/69/c8/a8994f5fece06db7c4a97c8fc1973684e178599b42e66280dded0524ef00/coverage-7.13.1-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:0403f647055de2609be776965108447deb8e384fe4a553c119e3ff6bfbab4784" },
{ url = "https://mirrors.aliyun.com/pypi/packages/cc/f7/91fa73c4b80305c86598a2d4e54ba22df6bf7d0d97500944af7ef155d9f7/coverage-7.13.1-cp313-cp313-win32.whl", hash = "sha256:549d195116a1ba1e1ae2f5ca143f9777800f6636eab917d4f02b5310d6d73461" },
{ url = "https://mirrors.aliyun.com/pypi/packages/45/0b/0768b4231d5a044da8f75e097a8714ae1041246bb765d6b5563bab456735/coverage-7.13.1-cp313-cp313-win_amd64.whl", hash = "sha256:5899d28b5276f536fcf840b18b61a9fce23cc3aec1d114c44c07fe94ebeaa500" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9b/b8/bdcb7253b7e85157282450262008f1366aa04663f3e3e4c30436f596c3e2/coverage-7.13.1-cp313-cp313-win_arm64.whl", hash = "sha256:868a2fae76dfb06e87291bcbd4dcbcc778a8500510b618d50496e520bd94d9b9" },
{ url = "https://mirrors.aliyun.com/pypi/packages/70/52/f2be52cc445ff75ea8397948c96c1b4ee14f7f9086ea62fc929c5ae7b717/coverage-7.13.1-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:67170979de0dacac3f3097d02b0ad188d8edcea44ccc44aaa0550af49150c7dc" },
{ url = "https://mirrors.aliyun.com/pypi/packages/47/79/c85e378eaa239e2edec0c5523f71542c7793fe3340954eafb0bc3904d32d/coverage-7.13.1-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:f80e2bb21bfab56ed7405c2d79d34b5dc0bc96c2c1d2a067b643a09fb756c43a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fe/9b/b1ade8bfb653c0bbce2d6d6e90cc6c254cbb99b7248531cc76253cb4da6d/coverage-7.13.1-cp313-cp313t-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:f83351e0f7dcdb14d7326c3d8d8c4e915fa685cbfdc6281f9470d97a04e9dfe4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1f/af/ebf91e3e1a2473d523e87e87fd8581e0aa08741b96265730e2d79ce78d8d/coverage-7.13.1-cp313-cp313t-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:bb3f6562e89bad0110afbe64e485aac2462efdce6232cdec7862a095dc3412f6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c4/8b/fb2423526d446596624ac7fde12ea4262e66f86f5120114c3cfd0bb2befa/coverage-7.13.1-cp313-cp313t-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:77545b5dcda13b70f872c3b5974ac64c21d05e65b1590b441c8560115dc3a0d1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9b/26/ef2adb1e22674913b89f0fe7490ecadcef4a71fa96f5ced90c60ec358789/coverage-7.13.1-cp313-cp313t-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:a4d240d260a1aed814790bbe1f10a5ff31ce6c21bc78f0da4a1e8268d6c80dbd" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ce/7d/f0f59b3404caf662e7b5346247883887687c074ce67ba453ea08c612b1d5/coverage-7.13.1-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:d2287ac9360dec3837bfdad969963a5d073a09a85d898bd86bea82aa8876ef3c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1a/b1/29896492b0b1a047604d35d6fa804f12818fa30cdad660763a5f3159e158/coverage-7.13.1-cp313-cp313t-musllinux_1_2_i686.whl", hash = "sha256:0d2c11f3ea4db66b5cbded23b20185c35066892c67d80ec4be4bab257b9ad1e0" },
{ url = "https://mirrors.aliyun.com/pypi/packages/48/f2/971de1238a62e6f0a4128d37adadc8bb882ee96afbe03ff1570291754629/coverage-7.13.1-cp313-cp313t-musllinux_1_2_riscv64.whl", hash = "sha256:3fc6a169517ca0d7ca6846c3c5392ef2b9e38896f61d615cb75b9e7134d4ee1e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/6a/fc/0474efcbb590ff8628830e9aaec5f1831594874360e3251f1fdec31d07a3/coverage-7.13.1-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:d10a2ed46386e850bb3de503a54f9fe8192e5917fcbb143bfef653a9355e9a53" },
{ url = "https://mirrors.aliyun.com/pypi/packages/88/4f/3c159b7953db37a7b44c0eab8a95c37d1aa4257c47b4602c04022d5cb975/coverage-7.13.1-cp313-cp313t-win32.whl", hash = "sha256:75a6f4aa904301dab8022397a22c0039edc1f51e90b83dbd4464b8a38dc87842" },
{ url = "https://mirrors.aliyun.com/pypi/packages/58/a5/6b57d28f81417f9335774f20679d9d13b9a8fb90cd6160957aa3b54a2379/coverage-7.13.1-cp313-cp313t-win_amd64.whl", hash = "sha256:309ef5706e95e62578cda256b97f5e097916a2c26247c287bbe74794e7150df2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/81/7c/160796f3b035acfbb58be80e02e484548595aa67e16a6345e7910ace0a38/coverage-7.13.1-cp313-cp313t-win_arm64.whl", hash = "sha256:92f980729e79b5d16d221038dbf2e8f9a9136afa072f9d5d6ed4cb984b126a09" },
{ url = "https://mirrors.aliyun.com/pypi/packages/cc/48/d9f421cb8da5afaa1a64570d9989e00fb7955e6acddc5a12979f7666ef60/coverage-7.13.1-py3-none-any.whl", hash = "sha256:2016745cb3ba554469d02819d78958b571792bb68e31302610e898f80dd3a573" },
Add ut test into CI (#11753) ### What problem does this PR solve? As title ### Type of change - [x] Other (please describe): --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-05 11:40:16 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "cramjam"
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
version = "2.11.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/14/12/34bf6e840a79130dfd0da7badfb6f7810b8fcfd60e75b0539372667b41b6/cramjam-2.11.0.tar.gz", hash = "sha256:5c82500ed91605c2d9781380b378397012e25127e89d64f460fea6aeac4389b4" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/0a/8b/406c5dc0f8e82385519d8c299c40fd6a56d97eca3fcd6f5da8dad48de75b/cramjam-2.11.0-cp313-cp313-macosx_10_12_x86_64.macosx_11_0_arm64.macosx_10_12_universal2.whl", hash = "sha256:2c289729cc1c04e88bafa48b51082fb462b0a57dbc96494eab2be9b14dca62af" },
{ url = "https://mirrors.aliyun.com/pypi/packages/00/ad/4186884083d6e4125b285903e17841827ab0d6d0cffc86216d27ed91e91d/cramjam-2.11.0-cp313-cp313-macosx_10_12_x86_64.whl", hash = "sha256:045201ee17147e36cf43d8ae2fa4b4836944ac672df5874579b81cf6d40f1a1f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/54/01/91b485cf76a7efef638151e8a7d35784dae2c4ff221b1aec2c083e4b106d/cramjam-2.11.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:619cd195d74c9e1d2a3ad78d63451d35379c84bd851aec552811e30842e1c67a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/cd/84/d0c80d279b2976870fc7d10f15dcb90a3c10c06566c6964b37c152694974/cramjam-2.11.0-cp313-cp313-manylinux_2_12_i686.manylinux2010_i686.whl", hash = "sha256:6eb3ae5ab72edb2ed68bdc0f5710f0a6cad7fd778a610ec2c31ee15e32d3921e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d6/70/88f2a5cb904281ed5d3c111b8f7d5366639817a5470f059bcd26833fc870/cramjam-2.11.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:df7da3f4b19e3078f9635f132d31b0a8196accb2576e3213ddd7a77f93317c20" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b2/06/cf5b02081132537d28964fb385fcef9ed9f8a017dd7d8c59d317e53ba50d/cramjam-2.11.0-cp313-cp313-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:57286b289cd557ac76c24479d8ecfb6c3d5b854cce54ccc7671f9a2f5e2a2708" },
{ url = "https://mirrors.aliyun.com/pypi/packages/57/27/63525087ed40a53d1867021b9c4858b80cc86274ffe7225deed067d88d92/cramjam-2.11.0-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:28952fbbf8b32c0cb7fa4be9bcccfca734bf0d0989f4b509dc7f2f70ba79ae06" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c3/ef/dbba082c6ebfb6410da4dd39a64e654d7194fcfd4567f85991a83fa4ec32/cramjam-2.11.0-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:78ed2e4099812a438b545dfbca1928ec825e743cd253bc820372d6ef8c3adff4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/35/ce/d902b9358a46a086938feae83b2251720e030f06e46006f4c1fc0ac9da20/cramjam-2.11.0-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7d9aecd5c3845d415bd6c9957c93de8d93097e269137c2ecb0e5a5256374bdc8" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e8/03/982f54553244b0afcbdb2ad2065d460f0ab05a72a96896a969a1ca136a1e/cramjam-2.11.0-cp313-cp313-musllinux_1_1_aarch64.whl", hash = "sha256:362fcf4d6f5e1242a4540812455f5a594949190f6fbc04f2ffbfd7ae0266d788" },
{ url = "https://mirrors.aliyun.com/pypi/packages/74/5f/748e54cdb665ec098ec519e23caacc65fc5ae58718183b071e33fc1c45b4/cramjam-2.11.0-cp313-cp313-musllinux_1_1_armv7l.whl", hash = "sha256:13240b3dea41b1174456cb9426843b085dc1a2bdcecd9ee2d8f65ac5703374b0" },
{ url = "https://mirrors.aliyun.com/pypi/packages/69/81/c4e6cb06ed69db0dc81f9a8b1dc74995ebd4351e7a1877143f7031ff2700/cramjam-2.11.0-cp313-cp313-musllinux_1_1_i686.whl", hash = "sha256:c54eed83726269594b9086d827decc7d2015696e31b99bf9b69b12d9063584fe" },
{ url = "https://mirrors.aliyun.com/pypi/packages/13/5b/966365523ce8290a08e163e3b489626c5adacdff2b3da9da1b0823dfb14e/cramjam-2.11.0-cp313-cp313-musllinux_1_1_x86_64.whl", hash = "sha256:f8195006fdd0fc0a85b19df3d64a3ef8a240e483ae1dfc7ac6a4316019eb5df2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3a/7d/7f8eb5c534b72b32c6eb79d74585bfee44a9a5647a14040bb65c31c2572d/cramjam-2.11.0-cp313-cp313-win32.whl", hash = "sha256:ccf30e3fe6d770a803dcdf3bb863fa44ba5dc2664d4610ba2746a3c73599f2e4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/37/05/47b5e0bf7c41a3b1cdd3b7c2147f880c93226a6bef1f5d85183040cbdece/cramjam-2.11.0-cp313-cp313-win_amd64.whl", hash = "sha256:ee36348a204f0a68b03400f4736224e9f61d1c6a1582d7f875c1ca56f0254268" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "crawl4ai"
build(deps): bump crawl4ai from 0.8.9 to 0.9.0 (#16470)
2026-06-30 09:34:48 +08:00
version = "0.9.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
Refa: migrate chat models to LiteLLM (#9394) ### What problem does this PR solve? All models pass the mock response tests, which means that if a model can return the correct response, everything should work as expected. However, not all models have been fully tested in a real environment, the real API_KEY. I suggest actively monitoring the refactored models over the coming period to ensure they work correctly and fixing them step by step, or waiting to merge until most have been tested in practical environment. ### Type of change - [x] Refactoring
2025-08-12 10:59:20 +08:00
{ name = "aiofiles" },
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
{ name = "aiohttp" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "aiosqlite" },
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
{ name = "alphashape" },
{ name = "anyio" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "beautifulsoup4" },
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
{ name = "brotli" },
{ name = "chardet" },
{ name = "click" },
{ name = "cssselect" },
{ name = "fake-useragent" },
{ name = "httpx", extra = ["http2"] },
{ name = "humanize" },
{ name = "lark" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "lxml" },
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
{ name = "nltk" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "numpy" },
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
{ name = "patchright" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "pillow" },
{ name = "playwright" },
fix: upgrade crawl4ai to 0.8.0 (CVE-2026-26217) (#15415) ## Summary Upgrade crawl4ai from 0.7.6 to 0.8.0 to fix CVE-2026-26217. ## Vulnerability | Field | Value | |-------|-------| | **ID** | CVE-2026-26217 | | **Severity** | CRITICAL | | **Scanner** | trivy | | **Rule** | `CVE-2026-26217` | | **File** | `uv.lock` | | **Assessment** | Likely exploitable | **Description**: Crawl4AI Has Local File Inclusion in Docker API via file:// URLs ## Evidence **Scanner confirmation**: trivy rule `CVE-2026-26217` flagged this pattern. **Production code**: This file is in the production codebase, not test-only code. ## Threat Model Context This is a web service - vulnerabilities in request handlers are directly exploitable by remote attackers. ## Changes - `pyproject.toml` - `uv.lock` ## Verification - [x] Build passes - [x] Scanner re-scan confirms fix - [x] LLM code review passed --- *This change addresses a pattern flagged by static analysis. The code path handles user-influenced input and the fix reduces the attack surface against both manual and automated exploitation.* --- *Automated security fix by [OrbisAI Security](https://orbisappsec.com)*
2026-05-29 19:08:41 +05:30
{ name = "playwright-stealth" },
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
{ name = "psutil" },
{ name = "pydantic" },
{ name = "pyopenssl" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "python-dotenv" },
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
{ name = "pyyaml" },
Refa: migrate chat models to LiteLLM (#9394) ### What problem does this PR solve? All models pass the mock response tests, which means that if a model can return the correct response, everything should work as expected. However, not all models have been fully tested in a real environment, the real API_KEY. I suggest actively monitoring the refactored models over the coming period to ensure they work correctly and fixing them step by step, or waiting to merge until most have been tested in practical environment. ### Type of change - [x] Refactoring
2025-08-12 10:59:20 +08:00
{ name = "rank-bm25" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "requests" },
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
{ name = "rich" },
{ name = "shapely" },
Refa: migrate chat models to LiteLLM (#9394) ### What problem does this PR solve? All models pass the mock response tests, which means that if a model can return the correct response, everything should work as expected. However, not all models have been fully tested in a real environment, the real API_KEY. I suggest actively monitoring the refactored models over the coming period to ensure they work correctly and fixing them step by step, or waiting to merge until most have been tested in practical environment. ### Type of change - [x] Refactoring
2025-08-12 10:59:20 +08:00
{ name = "snowballstemmer" },
{ name = "xxhash" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
build(deps): bump crawl4ai from 0.8.9 to 0.9.0 (#16470)
2026-06-30 09:34:48 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/52/37/9b288598bb0049b918eeb00f029592c748465c8a3e819461e95d8b5f9dab/crawl4ai-0.9.0.tar.gz", hash = "sha256:00c1c3516d9ca24a9b5004523ae67974ace8a5baadc6e92e295afca73f077153" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
build(deps): bump crawl4ai from 0.8.9 to 0.9.0 (#16470)
2026-06-30 09:34:48 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/c9/19/2d634f26d9ad1281874c503c1043174042fef9af631dc10798f4f22c1544/crawl4ai-0.9.0-py3-none-any.whl", hash = "sha256:1296072d9cd92e79144c6823ed6f27ab3966f16ce44e79fcbe86943b3435c366" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
[[package]]
name = "crc32c"
version = "2.8"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/e3/66/7e97aa77af7cf6afbff26e3651b564fe41932599bc2d3dce0b2f73d4829a/crc32c-2.8.tar.gz", hash = "sha256:578728964e59c47c356aeeedee6220e021e124b9d3e8631d95d9a5e5f06e261c" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/e8/d8/3ae227890b3be40955a7144106ef4dd97d6123a82c2a5310cdab58ca49d8/crc32c-2.8-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:36f1e03ee9e9c6938e67d3bcb60e36f260170aa5f37da1185e04ef37b56af395" },
{ url = "https://mirrors.aliyun.com/pypi/packages/bd/8b/178d3f987cd0e049b484615512d3f91f3d2caeeb8ff336bb5896ae317438/crc32c-2.8-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:b2f3226b94b85a8dd9b3533601d7a63e9e3e8edf03a8a169830ee8303a199aeb" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f2/a1/48145ae2545ebc0169d3283ebe882da580ea4606bfb67cf4ca922ac3cfc3/crc32c-2.8-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:6e08628bc72d5b6bc8e0730e8f142194b610e780a98c58cb6698e665cb885a5b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/06/4b/cf05ed9d934cc30e5ae22f97c8272face420a476090e736615d9a6b53de0/crc32c-2.8-cp313-cp313-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:086f64793c5ec856d1ab31a026d52ad2b895ac83d7a38fce557d74eb857f0a82" },
{ url = "https://mirrors.aliyun.com/pypi/packages/15/ab/4b04801739faf36345f6ba1920be5b1c70282fec52f8280afd3613fb13e2/crc32c-2.8-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:bcf72ee7e0135b3d941c34bb2c26c3fc6bc207106b49fd89aaafaeae223ae209" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a9/1b/6e38dde5bfd2ea69b7f2ab6ec229fcd972a53d39e2db4efe75c0ac0382ce/crc32c-2.8-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:8a717dd9c3fd777d9bc6603717eae172887d402c4ab589d124ebd0184a83f89e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ce/45/012176ffee90059ae8ec7131019c71724ea472aa63e72c0c8edbd1fad1d7/crc32c-2.8-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:0450bb845b3c3c7b9bdc0b4e95620ec9a40824abdc8c86d6285c919a90743c1a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f0/2b/f557629842f9dec2b3461cb3a0d854bb586ec45b814cea58b082c32f0dde/crc32c-2.8-cp313-cp313-win32.whl", hash = "sha256:765d220bfcbcffa6598ac11eb1e10af0ee4802b49fe126aa6bf79f8ddb9931d1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d0/db/fd0f698c15d1e21d47c64181a98290665a08fcbb3940cd559e9c15bda57e/crc32c-2.8-cp313-cp313-win_amd64.whl", hash = "sha256:171ff0260d112c62abcce29332986950a57bddee514e0a2418bfde493ea06bb3" },
{ url = "https://mirrors.aliyun.com/pypi/packages/db/b9/8e5d7054fe8e7eecab10fd0c8e7ffb01439417bdb6de1d66a81c38fc4a20/crc32c-2.8-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:b977a32a3708d6f51703c8557008f190aaa434d7347431efb0e86fcbe78c2a50" },
{ url = "https://mirrors.aliyun.com/pypi/packages/55/5f/cc926c70057a63cc0c98a3c8a896eb15fc7e74d3034eadd53c94917c6cc3/crc32c-2.8-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:7399b01db4adaf41da2fb36fe2408e75a8d82a179a9564ed7619412e427b26d6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a1/8a/0660c44a2dd2cb6ccbb529eb363b9280f5c766f1017bc8355ed8d695bd94/crc32c-2.8-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:4379f73f9cdad31958a673d11a332ec725ca71572401ca865867229f5f15e853" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f5/5a/6108d2dfc0fe33522ce83ba07aed4b22014911b387afa228808a278e27cd/crc32c-2.8-cp313-cp313t-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:2e68264555fab19bab08331550dab58573e351a63ed79c869d455edd3b0aa417" },
{ url = "https://mirrors.aliyun.com/pypi/packages/84/1e/c054f9e390090c197abf3d2936f4f9effaf0c6ee14569ae03d6ddf86958a/crc32c-2.8-cp313-cp313t-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:b48f2486727b8d0e7ccbae4a34cb0300498433d2a9d6b49cb13cb57c2e3f19cb" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c8/ad/1650e5c3341e4a485f800ea83116d72965030c5d48ccc168fcc685756e4d/crc32c-2.8-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:ecf123348934a086df8c8fde7f9f2d716d523ca0707c5a1367b8bb00d8134823" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d7/3b/f2ed924b177729cbb2ab30ca2902abff653c31d48c95e7b66717a9ca9fcc/crc32c-2.8-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:e636ac60f76de538f7a2c0d0f3abf43104ee83a8f5e516f6345dc283ed1a4df7" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4b/80/413b05ee6ace613208b31b3670c3135ee1cf451f0e72a9c839b4946acc04/crc32c-2.8-cp313-cp313t-win32.whl", hash = "sha256:8dd4a19505e0253892e1b2f1425cc3bd47f79ae5a04cb8800315d00aad7197f2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3b/1b/85eddb6ac5b38496c4e35c20298aae627970c88c3c624a22ab33e84f16c7/crc32c-2.8-cp313-cp313t-win_amd64.whl", hash = "sha256:4bb18e4bd98fb266596523ffc6be9c5b2387b2fa4e505ec56ca36336f49cb639" },
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
]
[[package]]
name = "crcmod"
version = "1.7"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/6b/b0/e595ce2a2527e169c3bcd6c33d2473c1918e0b7f6826a043ca1245dd4e5b/crcmod-1.7.tar.gz", hash = "sha256:dc7051a0db5f2bd48665a990d3ec1cc305a466a77358ca4492826f41f283601e" }
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "cryptography"
fix: upgrade crawl4ai to 0.8.0 (CVE-2026-26217) (#15415) ## Summary Upgrade crawl4ai from 0.7.6 to 0.8.0 to fix CVE-2026-26217. ## Vulnerability | Field | Value | |-------|-------| | **ID** | CVE-2026-26217 | | **Severity** | CRITICAL | | **Scanner** | trivy | | **Rule** | `CVE-2026-26217` | | **File** | `uv.lock` | | **Assessment** | Likely exploitable | **Description**: Crawl4AI Has Local File Inclusion in Docker API via file:// URLs ## Evidence **Scanner confirmation**: trivy rule `CVE-2026-26217` flagged this pattern. **Production code**: This file is in the production codebase, not test-only code. ## Threat Model Context This is a web service - vulnerabilities in request handlers are directly exploitable by remote attackers. ## Changes - `pyproject.toml` - `uv.lock` ## Verification - [x] Build passes - [x] Scanner re-scan confirms fix - [x] LLM code review passed --- *This change addresses a pattern flagged by static analysis. The code path handles user-influenced input and the fix reduces the attack surface against both manual and automated exploitation.* --- *Automated security fix by [OrbisAI Security](https://orbisappsec.com)*
2026-05-29 19:08:41 +05:30
version = "46.0.7"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "cffi", marker = "platform_python_implementation != 'PyPy'" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/47/93/ac8f3d5ff04d54bc814e961a43ae5b0b146154c89c61b47bb07557679b18/cryptography-46.0.7.tar.gz", hash = "sha256:e4cfd68c5f3e0bfdad0d38e023239b96a2fe84146481852dffbcca442c245aa5" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/0b/5d/4a8f770695d73be252331e60e526291e3df0c9b27556a90a6b47bccca4c2/cryptography-46.0.7-cp311-abi3-macosx_10_9_universal2.whl", hash = "sha256:ea42cbe97209df307fdc3b155f1b6fa2577c0defa8f1f7d3be7d31d189108ad4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/5f/45/6d80dc379b0bbc1f9d1e429f42e4cb9e1d319c7a8201beffd967c516ea01/cryptography-46.0.7-cp311-abi3-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:b36a4695e29fe69215d75960b22577197aca3f7a25b9cf9d165dcfe9d80bc325" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4a/9a/1765afe9f572e239c3469f2cb429f3ba7b31878c893b246b4b2994ffe2fe/cryptography-46.0.7-cp311-abi3-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:5ad9ef796328c5e3c4ceed237a183f5d41d21150f972455a9d926593a1dcb308" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8f/3e/af9246aaf23cd4ee060699adab1e47ced3f5f7e7a8ffdd339f817b446462/cryptography-46.0.7-cp311-abi3-manylinux_2_28_aarch64.whl", hash = "sha256:73510b83623e080a2c35c62c15298096e2a5dc8d51c3b4e1740211839d0dea77" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0f/54/6bbbfc5efe86f9d71041827b793c24811a017c6ac0fd12883e4caa86b8ed/cryptography-46.0.7-cp311-abi3-manylinux_2_28_ppc64le.whl", hash = "sha256:cbd5fb06b62bd0721e1170273d3f4d5a277044c47ca27ee257025146c34cbdd1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2d/cf/054b9d8220f81509939599c8bdbc0c408dbd2bdd41688616a20731371fe0/cryptography-46.0.7-cp311-abi3-manylinux_2_28_x86_64.whl", hash = "sha256:420b1e4109cc95f0e5700eed79908cef9268265c773d3a66f7af1eef53d409ef" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f9/46/4e4e9c6040fb01c7467d47217d2f882daddeb8828f7df800cb806d8a2288/cryptography-46.0.7-cp311-abi3-manylinux_2_31_armv7l.whl", hash = "sha256:24402210aa54baae71d99441d15bb5a1919c195398a87b563df84468160a65de" },
{ url = "https://mirrors.aliyun.com/pypi/packages/36/5f/313586c3be5a2fbe87e4c9a254207b860155a8e1f3cca99f9910008e7d08/cryptography-46.0.7-cp311-abi3-manylinux_2_34_aarch64.whl", hash = "sha256:8a469028a86f12eb7d2fe97162d0634026d92a21f3ae0ac87ed1c4a447886c83" },
{ url = "https://mirrors.aliyun.com/pypi/packages/69/33/60dfc4595f334a2082749673386a4d05e4f0cf4df8248e63b2c3437585f2/cryptography-46.0.7-cp311-abi3-manylinux_2_34_ppc64le.whl", hash = "sha256:9694078c5d44c157ef3162e3bf3946510b857df5a3955458381d1c7cfc143ddb" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c7/0b/333ddab4270c4f5b972f980adef4faa66951a4aaf646ca067af597f15563/cryptography-46.0.7-cp311-abi3-manylinux_2_34_x86_64.whl", hash = "sha256:42a1e5f98abb6391717978baf9f90dc28a743b7d9be7f0751a6f56a75d14065b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d2/14/633913398b43b75f1234834170947957c6b623d1701ffc7a9600da907e89/cryptography-46.0.7-cp311-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:91bbcb08347344f810cbe49065914fe048949648f6bd5c2519f34619142bbe85" },
{ url = "https://mirrors.aliyun.com/pypi/packages/10/f2/19ceb3b3dc14009373432af0c13f46aa08e3ce334ec6eff13492e1812ccd/cryptography-46.0.7-cp311-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:5d1c02a14ceb9148cc7816249f64f623fbfee39e8c03b3650d842ad3f34d637e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1a/bb/a5c213c19ee94b15dfccc48f363738633a493812687f5567addbcbba9f6f/cryptography-46.0.7-cp311-abi3-win32.whl", hash = "sha256:d23c8ca48e44ee015cd0a54aeccdf9f09004eba9fc96f38c911011d9ff1bd457" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2b/02/7788f9fefa1d060ca68717c3901ae7fffa21ee087a90b7f23c7a603c32ae/cryptography-46.0.7-cp311-abi3-win_amd64.whl", hash = "sha256:397655da831414d165029da9bc483bed2fe0e75dde6a1523ec2fe63f3c46046b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a7/7f/cd42fc3614386bc0c12f0cb3c4ae1fc2bbca5c9662dfed031514911d513d/cryptography-46.0.7-cp38-abi3-macosx_10_9_universal2.whl", hash = "sha256:462ad5cb1c148a22b2e3bcc5ad52504dff325d17daf5df8d88c17dda1f75f2a4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a5/d0/36a49f0262d2319139d2829f773f1b97ef8aef7f97e6e5bd21455e5a8fb5/cryptography-46.0.7-cp38-abi3-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:84d4cced91f0f159a7ddacad249cc077e63195c36aac40b4150e7a57e84fffe7" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8a/6c/1a42450f464dda6ffbe578a911f773e54dd48c10f9895a23a7e88b3e7db5/cryptography-46.0.7-cp38-abi3-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:128c5edfe5e5938b86b03941e94fac9ee793a94452ad1365c9fc3f4f62216832" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9a/92/4ed714dbe93a066dc1f4b4581a464d2d7dbec9046f7c8b7016f5286329e2/cryptography-46.0.7-cp38-abi3-manylinux_2_28_aarch64.whl", hash = "sha256:5e51be372b26ef4ba3de3c167cd3d1022934bc838ae9eaad7e644986d2a3d163" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b7/e6/a26b84096eddd51494bba19111f8fffe976f6a09f132706f8f1bf03f51f7/cryptography-46.0.7-cp38-abi3-manylinux_2_28_ppc64le.whl", hash = "sha256:cdf1a610ef82abb396451862739e3fc93b071c844399e15b90726ef7470eeaf2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c7/08/ffd537b605568a148543ac3c2b239708ae0bd635064bab41359252ef88ed/cryptography-46.0.7-cp38-abi3-manylinux_2_28_x86_64.whl", hash = "sha256:1d25aee46d0c6f1a501adcddb2d2fee4b979381346a78558ed13e50aa8a59067" },
{ url = "https://mirrors.aliyun.com/pypi/packages/16/01/0cd51dd86ab5b9befe0d031e276510491976c3a80e9f6e31810cce46c4ad/cryptography-46.0.7-cp38-abi3-manylinux_2_31_armv7l.whl", hash = "sha256:cdfbe22376065ffcf8be74dc9a909f032df19bc58a699456a21712d6e5eabfd0" },
{ url = "https://mirrors.aliyun.com/pypi/packages/92/49/819d6ed3a7d9349c2939f81b500a738cb733ab62fbecdbc1e38e83d45e12/cryptography-46.0.7-cp38-abi3-manylinux_2_34_aarch64.whl", hash = "sha256:abad9dac36cbf55de6eb49badd4016806b3165d396f64925bf2999bcb67837ba" },
{ url = "https://mirrors.aliyun.com/pypi/packages/80/07/ad9b3c56ebb95ed2473d46df0847357e01583f4c52a85754d1a55e29e4d0/cryptography-46.0.7-cp38-abi3-manylinux_2_34_ppc64le.whl", hash = "sha256:935ce7e3cfdb53e3536119a542b839bb94ec1ad081013e9ab9b7cfd478b05006" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b8/c7/201d3d58f30c4c2bdbe9b03844c291feb77c20511cc3586daf7edc12a47b/cryptography-46.0.7-cp38-abi3-manylinux_2_34_x86_64.whl", hash = "sha256:35719dc79d4730d30f1c2b6474bd6acda36ae2dfae1e3c16f2051f215df33ce0" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a5/ef/649750cbf96f3033c3c976e112265c33906f8e462291a33d77f90356548c/cryptography-46.0.7-cp38-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:7bbc6ccf49d05ac8f7d7b5e2e2c33830d4fe2061def88210a126d130d7f71a85" },
{ url = "https://mirrors.aliyun.com/pypi/packages/41/52/a8908dcb1a389a459a29008c29966c1d552588d4ae6d43f3a1a4512e0ebe/cryptography-46.0.7-cp38-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:a1529d614f44b863a7b480c6d000fe93b59acee9c82ffa027cfadc77521a9f5e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4b/fa/f0ab06238e899cc3fb332623f337a7364f36f4bb3f2534c2bb95a35b132c/cryptography-46.0.7-cp38-abi3-win32.whl", hash = "sha256:f247c8c1a1fb45e12586afbb436ef21ff1e80670b2861a90353d9b025583d246" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d2/f1/00ce3bde3ca542d1acd8f8cfa38e446840945aa6363f9b74746394b14127/cryptography-46.0.7-cp38-abi3-win_amd64.whl", hash = "sha256:506c4ff91eff4f82bdac7633318a526b1d1309fc07ca76a3ad182cb5b686d6d3" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "cssselect"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.3.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/72/0a/c3ea9573b1dc2e151abfe88c7fe0c26d1892fe6ed02d0cdb30f0d57029d5/cssselect-1.3.0.tar.gz", hash = "sha256:57f8a99424cfab289a1b6a816a43075a4b00948c86b4dcf3ef4ee7e15f7ab0c7" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ee/58/257350f7db99b4ae12b614a36256d9cc870d71d9e451e79c2dc3b23d7c3c/cssselect-1.3.0-py3-none-any.whl", hash = "sha256:56d1bf3e198080cc1667e137bc51de9cadfca259f03c2d4e09037b3e01e30f0d" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
[[package]]
name = "curl-cffi"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "0.14.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
dependencies = [
{ name = "certifi" },
{ name = "cffi" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/9b/c9/0067d9a25ed4592b022d4558157fcdb6e123516083700786d38091688767/curl_cffi-0.14.0.tar.gz", hash = "sha256:5ffbc82e59f05008ec08ea432f0e535418823cda44178ee518906a54f27a5f0f" }
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/aa/f0/0f21e9688eaac85e705537b3a87a5588d0cefb2f09d83e83e0e8be93aa99/curl_cffi-0.14.0-cp39-abi3-macosx_14_0_arm64.whl", hash = "sha256:e35e89c6a69872f9749d6d5fda642ed4fc159619329e99d577d0104c9aad5893" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ba/a3/0419bd48fce5b145cb6a2344c6ac17efa588f5b0061f212c88e0723da026/curl_cffi-0.14.0-cp39-abi3-macosx_15_0_x86_64.whl", hash = "sha256:5945478cd28ad7dfb5c54473bcfb6743ee1d66554d57951fdf8fc0e7d8cf4e45" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e2/07/a238dd062b7841b8caa2fa8a359eb997147ff3161288f0dd46654d898b4d/curl_cffi-0.14.0-cp39-abi3-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:c42e8fa3c667db9ccd2e696ee47adcd3cd5b0838d7282f3fc45f6c0ef3cfdfa7" },
{ url = "https://mirrors.aliyun.com/pypi/packages/7c/d2/ce907c9b37b5caf76ac08db40cc4ce3d9f94c5500db68a195af3513eacbc/curl_cffi-0.14.0-cp39-abi3-manylinux_2_26_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:060fe2c99c41d3cb7f894de318ddf4b0301b08dca70453d769bd4e74b36b8483" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f2/ae/6256995b18c75e6ef76b30753a5109e786813aa79088b27c8eabb1ef85c9/curl_cffi-0.14.0-cp39-abi3-manylinux_2_28_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:b158c41a25388690dd0d40b5bc38d1e0f512135f17fdb8029868cbc1993d2e5b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fb/10/ff64249e516b103cb762e0a9dca3ee0f04cf25e2a1d5d9838e0f1273d071/curl_cffi-0.14.0-cp39-abi3-manylinux_2_28_i686.whl", hash = "sha256:1439fbef3500fb723333c826adf0efb0e2e5065a703fb5eccce637a2250db34a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/51/76/d6f7bb76c2d12811aa7ff16f5e17b678abdd1b357b9a8ac56310ceccabd5/curl_cffi-0.14.0-cp39-abi3-manylinux_2_34_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:e7176f2c2d22b542e3cf261072a81deb018cfa7688930f95dddef215caddb469" },
{ url = "https://mirrors.aliyun.com/pypi/packages/23/7c/cca39c0ed4e1772613d3cba13091c0e9d3b89365e84b9bf9838259a3cd8f/curl_cffi-0.14.0-cp39-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:03f21ade2d72978c2bb8670e9b6de5260e2755092b02d94b70b906813662998d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/75/03/a942d7119d3e8911094d157598ae0169b1c6ca1bd3f27d7991b279bcc45b/curl_cffi-0.14.0-cp39-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:58ebf02de64ee5c95613209ddacb014c2d2f86298d7080c0a1c12ed876ee0690" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a2/77/78900e9b0833066d2274bda75cba426fdb4cef7fbf6a4f6a6ca447607bec/curl_cffi-0.14.0-cp39-abi3-win_amd64.whl", hash = "sha256:6e503f9a103f6ae7acfb3890c843b53ec030785a22ae7682a22cc43afb94123e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/5c/7c/d2ba86b0b3e1e2830bd94163d047de122c69a8df03c5c7c36326c456ad82/curl_cffi-0.14.0-cp39-abi3-win_arm64.whl", hash = "sha256:2eed50a969201605c863c4c31269dfc3e0da52916086ac54553cfa353022425c" },
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "cycler"
version = "0.12.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a9/95/a3dbbb5028f35eafb79008e7522a75244477d2838f38cbb722248dabc2a8/cycler-0.12.1.tar.gz", hash = "sha256:88bb128f02ba341da8ef447245a9e138fae777f6a23943da4540077d3601eb1c" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/e7/05/c19819d5e3d95294a6f5947fb9b9629efb316b96de511b418c53d245aae6/cycler-0.12.1-py3-none-any.whl", hash = "sha256:85cef7cff222d8644161529808465972e51340599459b8ac3ccbac5a854e0d30" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
[[package]]
name = "cymem"
version = "2.0.13"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/c0/8f/2f0fbb32535c3731b7c2974c569fb9325e0a38ed5565a08e1139a3b71e82/cymem-2.0.13.tar.gz", hash = "sha256:1c91a92ae8c7104275ac26bd4d29b08ccd3e7faff5893d3858cb6fadf1bc1588" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/ce/0f/95a4d1e3bebfdfa7829252369357cf9a764f67569328cd9221f21e2c952e/cymem-2.0.13-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:891fd9030293a8b652dc7fb9fdc79a910a6c76fc679cd775e6741b819ffea476" },
{ url = "https://mirrors.aliyun.com/pypi/packages/bf/a0/8fc929cc29ae466b7b4efc23ece99cbd3ea34992ccff319089c624d667fd/cymem-2.0.13-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:89c4889bd16513ce1644ccfe1e7c473ba7ca150f0621e66feac3a571bde09e7e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4a/b3/deeb01354ebaf384438083ffe0310209ef903db3e7ba5a8f584b06d28387/cymem-2.0.13-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:45dcaba0f48bef9cc3d8b0b92058640244a95a9f12542210b51318da97c2cf28" },
{ url = "https://mirrors.aliyun.com/pypi/packages/36/36/bc980b9a14409f3356309c45a8d88d58797d02002a9d794dd6c84e809d3a/cymem-2.0.13-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:e96848faaafccc0abd631f1c5fb194eac0caee4f5a8777fdbb3e349d3a21741c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fd/dd/a12522952624685bd0f8968e26d2ed6d059c967413ce6eb52292f538f1b0/cymem-2.0.13-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:e02d3e2c3bfeb21185d5a4a70790d9df40629a87d8d7617dc22b4e864f665fa3" },
{ url = "https://mirrors.aliyun.com/pypi/packages/08/11/5dc933ddfeb2dfea747a0b935cb965b9a7580b324d96fc5f5a1b5ff8df29/cymem-2.0.13-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:fece5229fd5ecdcd7a0738affb8c59890e13073ae5626544e13825f26c019d3c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/70/66/d23b06166864fa94e13a98e5922986ce774832936473578febce64448d75/cymem-2.0.13-cp313-cp313-win_amd64.whl", hash = "sha256:38aefeb269597c1a0c2ddf1567dd8605489b661fa0369c6406c1acd433b4c7ba" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2f/9e/c7b21271ab88a21760f3afdec84d2bc09ffa9e6c8d774ad9d4f1afab0416/cymem-2.0.13-cp313-cp313-win_arm64.whl", hash = "sha256:717270dcfd8c8096b479c42708b151002ff98e434a7b6f1f916387a6c791e2ad" },
{ url = "https://mirrors.aliyun.com/pypi/packages/7f/28/d3b03427edc04ae04910edf1c24b993881c3ba93a9729a42bcbb816a1808/cymem-2.0.13-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:7e1a863a7f144ffb345397813701509cfc74fc9ed360a4d92799805b4b865dd1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/35/a9/7ed53e481f47ebfb922b0b42e980cec83e98ccb2137dc597ea156642440c/cymem-2.0.13-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:c16cb80efc017b054f78998c6b4b013cef509c7b3d802707ce1f85a1d68361bf" },
{ url = "https://mirrors.aliyun.com/pypi/packages/61/39/a3d6ad073cf7f0fbbb8bbf09698c3c8fac11be3f791d710239a4e8dd3438/cymem-2.0.13-cp313-cp313t-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:0d78a27c88b26c89bd1ece247d1d5939dba05a1dae6305aad8fd8056b17ddb51" },
{ url = "https://mirrors.aliyun.com/pypi/packages/36/0c/20697c8bc19f624a595833e566f37d7bcb9167b0ce69de896eba7cfc9c2d/cymem-2.0.13-cp313-cp313t-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:6d36710760f817194dacb09d9fc45cb6a5062ed75e85f0ef7ad7aeeb13d80cc3" },
{ url = "https://mirrors.aliyun.com/pypi/packages/82/d4/9326e3422d1c2d2b4a8fb859bdcce80138f6ab721ddafa4cba328a505c71/cymem-2.0.13-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:c8f30971cadd5dcf73bcfbbc5849b1f1e1f40db8cd846c4aa7d3b5e035c7b583" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ed/bc/68da7dd749b72884dc22e898562f335002d70306069d496376e5ff3b6153/cymem-2.0.13-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:9d441d0e45798ec1fd330373bf7ffa6b795f229275f64016b6a193e6e2a51522" },
{ url = "https://mirrors.aliyun.com/pypi/packages/50/23/dbf2ad6ecd19b99b3aab6203b1a06608bbd04a09c522d836b854f2f30f73/cymem-2.0.13-cp313-cp313t-win_amd64.whl", hash = "sha256:d1c950eebb9f0f15e3ef3591313482a5a611d16fc12d545e2018cd607f40f472" },
{ url = "https://mirrors.aliyun.com/pypi/packages/54/3f/35701c13e1fc7b0895198c8b20068c569a841e0daf8e0b14d1dc0816b28f/cymem-2.0.13-cp313-cp313t-win_arm64.whl", hash = "sha256:042e8611ef862c34a97b13241f5d0da86d58aca3cecc45c533496678e75c5a1f" },
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
]
fix: upgrade crawl4ai to 0.8.0 (CVE-2026-26217) (#15415) ## Summary Upgrade crawl4ai from 0.7.6 to 0.8.0 to fix CVE-2026-26217. ## Vulnerability | Field | Value | |-------|-------| | **ID** | CVE-2026-26217 | | **Severity** | CRITICAL | | **Scanner** | trivy | | **Rule** | `CVE-2026-26217` | | **File** | `uv.lock` | | **Assessment** | Likely exploitable | **Description**: Crawl4AI Has Local File Inclusion in Docker API via file:// URLs ## Evidence **Scanner confirmation**: trivy rule `CVE-2026-26217` flagged this pattern. **Production code**: This file is in the production codebase, not test-only code. ## Threat Model Context This is a web service - vulnerabilities in request handlers are directly exploitable by remote attackers. ## Changes - `pyproject.toml` - `uv.lock` ## Verification - [x] Build passes - [x] Scanner re-scan confirms fix - [x] LLM code review passed --- *This change addresses a pattern flagged by static analysis. The code path handles user-influenced input and the fix reduces the attack surface against both manual and automated exploitation.* --- *Automated security fix by [OrbisAI Security](https://orbisappsec.com)*
2026-05-29 19:08:41 +05:30
[[package]]
name = "cython"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "3.2.5"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/3f/3b/ebd94c8b85f8e41b5015a9ed94ee3df866024d480d05cd08b774684fb81d/cython-3.2.5.tar.gz", hash = "sha256:3dd42e4cf36ad15f265bdfec2337cc00c688c8eb6d374ffd13bb19437c27bba1" }
fix: upgrade crawl4ai to 0.8.0 (CVE-2026-26217) (#15415) ## Summary Upgrade crawl4ai from 0.7.6 to 0.8.0 to fix CVE-2026-26217. ## Vulnerability | Field | Value | |-------|-------| | **ID** | CVE-2026-26217 | | **Severity** | CRITICAL | | **Scanner** | trivy | | **Rule** | `CVE-2026-26217` | | **File** | `uv.lock` | | **Assessment** | Likely exploitable | **Description**: Crawl4AI Has Local File Inclusion in Docker API via file:// URLs ## Evidence **Scanner confirmation**: trivy rule `CVE-2026-26217` flagged this pattern. **Production code**: This file is in the production codebase, not test-only code. ## Threat Model Context This is a web service - vulnerabilities in request handlers are directly exploitable by remote attackers. ## Changes - `pyproject.toml` - `uv.lock` ## Verification - [x] Build passes - [x] Scanner re-scan confirms fix - [x] LLM code review passed --- *This change addresses a pattern flagged by static analysis. The code path handles user-influenced input and the fix reduces the attack surface against both manual and automated exploitation.* --- *Automated security fix by [OrbisAI Security](https://orbisappsec.com)*
2026-05-29 19:08:41 +05:30
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/7e/30/f648409de61fd74ae63090071061145059664cc9b9ff8578197601a3beb6/cython-3.2.5-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:6e5d7a60835345a8bd29d3aa57070880cc3ce017ea0ade7b9f771ce4bf539b1f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a3/de/e3e0cf5704fe569d54b8cd5dc316c9fbf08b1b74728732f86e90168b7a3f/cython-3.2.5-cp39-abi3-macosx_10_9_x86_64.whl", hash = "sha256:224149d18d980e6ea5001b70fc7ce096c1891d59035dfa9cc5ede50f55804913" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d4/5c/9cd909e6a8bb178e4e0f9a2a9227c8201a2be38abe45ada4a4c3e9154277/cython-3.2.5-py3-none-any.whl", hash = "sha256:dc1c8cebb7df5bce37f5f8dc1e5bf04313272a5973d50a55c0ec76c83812911b" },
fix: upgrade crawl4ai to 0.8.0 (CVE-2026-26217) (#15415) ## Summary Upgrade crawl4ai from 0.7.6 to 0.8.0 to fix CVE-2026-26217. ## Vulnerability | Field | Value | |-------|-------| | **ID** | CVE-2026-26217 | | **Severity** | CRITICAL | | **Scanner** | trivy | | **Rule** | `CVE-2026-26217` | | **File** | `uv.lock` | | **Assessment** | Likely exploitable | **Description**: Crawl4AI Has Local File Inclusion in Docker API via file:// URLs ## Evidence **Scanner confirmation**: trivy rule `CVE-2026-26217` flagged this pattern. **Production code**: This file is in the production codebase, not test-only code. ## Threat Model Context This is a web service - vulnerabilities in request handlers are directly exploitable by remote attackers. ## Changes - `pyproject.toml` - `uv.lock` ## Verification - [x] Build passes - [x] Scanner re-scan confirms fix - [x] LLM code review passed --- *This change addresses a pattern flagged by static analysis. The code path handles user-influenced input and the fix reduces the attack surface against both manual and automated exploitation.* --- *Automated security fix by [OrbisAI Security](https://orbisappsec.com)*
2026-05-29 19:08:41 +05:30
]
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
[[package]]
name = "darabonba-core"
version = "1.0.5"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
dependencies = [
{ name = "aiohttp" },
{ name = "alibabacloud-tea" },
{ name = "requests" },
]
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/66/d3/a7daaee544c904548e665829b51a9fa2572acb82c73ad787a8ff90273002/darabonba_core-1.0.5-py3-none-any.whl", hash = "sha256:671ab8dbc4edc2a8f88013da71646839bb8914f1259efc069353243ef52ea27c" },
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "dashscope"
Chore/upgrade dashscope to 1.25.11 (#13007) ## Description Upgrade dashscope package to support text-embedding-v4 model. ## Changes - Update dashscope version from 1.20.11 to 1.25.11 in pyproject.toml ## Reason The text-embedding-v4 model requires dashscope >= 1.25.0 to function properly. This upgrade ensures compatibility with the latest embedding models. Co-authored-by: Clint-chan <Clint-chan@users.noreply.github.com>
2026-02-06 19:06:41 +08:00
version = "1.25.11"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "aiohttp" },
Chore/upgrade dashscope to 1.25.11 (#13007) ## Description Upgrade dashscope package to support text-embedding-v4 model. ## Changes - Update dashscope version from 1.20.11 to 1.25.11 in pyproject.toml ## Reason The text-embedding-v4 model requires dashscope >= 1.25.0 to function properly. This upgrade ensures compatibility with the latest embedding models. Co-authored-by: Clint-chan <Clint-chan@users.noreply.github.com>
2026-02-06 19:06:41 +08:00
{ name = "certifi" },
{ name = "cryptography" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "requests" },
{ name = "websocket-client" },
]
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/30/15/35551e6c6d3ea19df754ed32aa5f281b2052ef9e1ff1538f2708f74f3312/dashscope-1.25.11-py3-none-any.whl", hash = "sha256:93e86437f5f30e759e98292f0490e44eff00c337968363f27d29dd42ec7cc07c" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "datrie"
Don't release full image (#10654) ### What problem does this PR solve? Introduced gpu profile in .env Added Dockerfile_tei fix datrie Removed LIGHTEN flag ### Type of change - [x] Documentation Update - [x] Refactoring
2025-10-23 23:02:27 +08:00
version = "0.8.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/51/0b/c0f53a14317b304e2e93b29a831b0c83306caae9af7f0e2e037d17c4f63f/datrie-0.8.3.tar.gz", hash = "sha256:ea021ad4c8a8bf14e08a71c7872a622aa399a510f981296825091c7ca0436e80" }
Code refactor (#5371) ### What problem does this PR solve? #5173 ### Type of change - [x] Refactoring
2025-02-26 15:40:52 +08:00
Go: add balance command (#14262) ### What problem does this PR solve? ``` RAGFlow(user)> list supported models from 'moonshot' 'test'; +---------------------------------+ | model_name | +---------------------------------+ | moonshot-v1-32k-vision-preview | | kimi-k2.6 | | moonshot-v1-8k | | moonshot-v1-auto | | moonshot-v1-128k | | moonshot-v1-32k | | kimi-k2.5 | | moonshot-v1-8k-vision-preview | | moonshot-v1-128k-vision-preview | +---------------------------------+ RAGFlow(user)> show balance from 'moonshot' 'test'; +---------+----------+ | balance | currency | +---------+----------+ | 0 | CNY | +---------+----------+ ``` ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2026-04-21 21:31:50 +08:00
[[package]]
name = "debugpy"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.8.21"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f2/aa/12037145b7a56eaa5b29b41872f7a21b538e807e13f32c4d3c46e59be084/debugpy-1.8.21.tar.gz", hash = "sha256:a3c53278e84c94e11bd87c53970ec391d1a67396c8b22609fcac576520e611a6" }
Go: add balance command (#14262) ### What problem does this PR solve? ``` RAGFlow(user)> list supported models from 'moonshot' 'test'; +---------------------------------+ | model_name | +---------------------------------+ | moonshot-v1-32k-vision-preview | | kimi-k2.6 | | moonshot-v1-8k | | moonshot-v1-auto | | moonshot-v1-128k | | moonshot-v1-32k | | kimi-k2.5 | | moonshot-v1-8k-vision-preview | | moonshot-v1-128k-vision-preview | +---------------------------------+ RAGFlow(user)> show balance from 'moonshot' 'test'; +---------+----------+ | balance | currency | +---------+----------+ | 0 | CNY | +---------+----------+ ``` ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2026-04-21 21:31:50 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/77/6b/d817e1f8cc77aa055d37fba092e0febfdff40fe652d8d53d4cd7a86ad98d/debugpy-1.8.21-cp313-cp313-macosx_15_0_universal2.whl", hash = "sha256:13678151fc401e2d68c9880b91e28714f797d40422994572b24560ef80910a88" },
{ url = "https://mirrors.aliyun.com/pypi/packages/48/57/412421516afc3055fa577516f00beec3d663f9b0ab330639547ae6c57720/debugpy-1.8.21-cp313-cp313-manylinux_2_34_x86_64.whl", hash = "sha256:ecbd158386c31ffe71d46f72d44d56e66331ab9b16cad649156d514368f23ab2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c1/62/2c616337cf6ba7b07ebbc97f02c6c945a8e2f76b365e33ee809c32ee36d1/debugpy-1.8.21-cp313-cp313-win32.whl", hash = "sha256:2c2ae706dec41d99a9ca1f7ebc987a83e65578363be6f6b3ac9067504917fae1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f8/99/9175103392f84c4b1bf7622888cdc68da07f0ff7d9e581266428f6776033/debugpy-1.8.21-cp313-cp313-win_amd64.whl", hash = "sha256:aa648733047443eb1d07682c4ef287d36a54507b643ffdf38b09a3ef002c72a0" },
{ url = "https://mirrors.aliyun.com/pypi/packages/95/51/67e7cf11a53e40694f720457d5b3a1cdaaa3d5a9a633e482f225456b93ff/debugpy-1.8.21-py2.py3-none-any.whl", hash = "sha256:b1e37d333663c8851516a47364ef473da127f9caebe4417e6df6f5825a7e9a92" },
Go: add balance command (#14262) ### What problem does this PR solve? ``` RAGFlow(user)> list supported models from 'moonshot' 'test'; +---------------------------------+ | model_name | +---------------------------------+ | moonshot-v1-32k-vision-preview | | kimi-k2.6 | | moonshot-v1-8k | | moonshot-v1-auto | | moonshot-v1-128k | | moonshot-v1-32k | | kimi-k2.5 | | moonshot-v1-8k-vision-preview | | moonshot-v1-128k-vision-preview | +---------------------------------+ RAGFlow(user)> show balance from 'moonshot' 'test'; +---------+----------+ | balance | currency | +---------+----------+ | 0 | CNY | +---------+----------+ ``` ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2026-04-21 21:31:50 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "decorator"
Add tavily as web searh tool. (#5349) ### What problem does this PR solve? #5198 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-02-26 10:21:04 +08:00
version = "5.2.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/43/fa/6d96a0978d19e17b68d634497769987b16c8f4cd0a7a05048bec693caa6b/decorator-5.2.1.tar.gz", hash = "sha256:65f266143752f734b0a7cc83c46f4618af75b8c5911b00ccb61d0ac9b6da0360" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/4e/8c/f3147f5c4b73e7550fe5f9352eaa956ae838d5c51eb58e7a25b9f3e2643b/decorator-5.2.1-py3-none-any.whl", hash = "sha256:d316bb415a2d9e2d2b3abcc4084c6502fc09240e292cd76a76afc106a1c8e04a" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "deepl"
version = "1.18.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "requests" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/82/46/1dfe623c24aac5a341bf7eedabb8d1f719df8fd3a6f45aefcd0b83e96ce0/deepl-1.18.0.tar.gz", hash = "sha256:5ae41763939441edbca7640fd344280cbee47d490641ce35206910a8b01e778e" }
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/d3/02/fe0df955618c204ea51a2cee85461e736a939cf9d593b314adeb1c2c5d2e/deepl-1.18.0-py3-none-any.whl", hash = "sha256:2afe9adc459f5c591282e4d74570a0dc5041554d54dd687f72d3b0b77936e9ce" },
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
]
[[package]]
name = "defusedxml"
version = "0.7.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/0f/d5/c66da9b79e5bdb124974bfe172b4daf3c984ebd9c2a06e2b8a4dc7331c72/defusedxml-0.7.1.tar.gz", hash = "sha256:1bb3032db185915b62d7c6209c5a8792be6a32ab2fedacc84e01b52c51aa3e69" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/07/6c/aa3f2f849e01cb6a001cd8554a88d4c77c5c1a31c95bdf1cf9301e6d9ef4/defusedxml-0.7.1-py2.py3-none-any.whl", hash = "sha256:a352e7e428770286cc899e2542b6cdaedb2b4953ff269a210103ec58f6198a61" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "demjson3"
version = "3.0.6"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f7/d2/6a81a9b5311d50542e11218b470dafd8adbaf1b3e51fc1fddd8a57eed691/demjson3-3.0.6.tar.gz", hash = "sha256:37c83b0c6eb08d25defc88df0a2a4875d58a7809a9650bd6eee7afd8053cdbac" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "deprecated"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "1.3.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "wrapt" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/49/85/12f0a49a7c4ffb70572b6c2ef13c90c88fd190debda93b23f026b25f9634/deprecated-1.3.1.tar.gz", hash = "sha256:b1b50e0ff0c1fddaa5708a2c6b0a6588bb09b892825ab2b214ac9ea9d92a5223" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/84/d0/205d54408c08b13550c733c4b85429e7ead111c7f0014309637425520a9a/deprecated-1.3.1-py2.py3-none-any.whl", hash = "sha256:597bfef186b6f60181535a29fbe44865ce137a5079f295b479886c82729d5f3f" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "dill"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "0.4.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/12/80/630b4b88364e9a8c8c5797f4602d0f76ef820909ee32f0bacb9f90654042/dill-0.4.0.tar.gz", hash = "sha256:0633f1d2df477324f53a895b02c901fb961bdbf65a17122586ea7019292cbcf0" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/50/3d/9373ad9c56321fdab5b41197068e1d8c25883b3fea29dd361f9b55116869/dill-0.4.0-py3-none-any.whl", hash = "sha256:44f54bf6412c2c8464c14e8243eb163690a9800dbe2c367330883b19c7561049" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "discord-py"
version = "2.3.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "aiohttp" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/6f/cb/a360101905102684a4fe6fc543976842383f54ddeeef020959e4965c416e/discord.py-2.3.2.tar.gz", hash = "sha256:4560f70f2eddba7e83370ecebd237ac09fbb4980dc66507482b0c0e5b8f76b9c" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/9c/7e/5f1b24b2ced0c4b3042204f7827b57c7dcb26d368e9b0fde8cec7853cf30/discord.py-2.3.2-py3-none-any.whl", hash = "sha256:9da4679fc3cb10c64b388284700dc998663e0e57328283bbfcfc2525ec5960a6" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "diskcache"
version = "5.6.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/3f/21/1c1ffc1a039ddcc459db43cc108658f32c57d271d7289a2794e401d0fdb6/diskcache-5.6.3.tar.gz", hash = "sha256:2c3a3fa2743d8535d832ec61c2054a1641f41775aa7c556758a109941e33e4fc" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/3f/27/4570e78fc0bf5ea0ca45eb1de3818a23787af9b390c0b0a0033a1b8236f9/diskcache-5.6.3-py3-none-any.whl", hash = "sha256:5e31b2d5fbad117cc363ebaf6b689474db18a1f6438bc82358b024abd4c2ca19" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "distro"
version = "1.9.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/fc/f8/98eea607f65de6527f8a2e8885fc8015d3e6f5775df186e443e0964a11c3/distro-1.9.0.tar.gz", hash = "sha256:2fa77c6fd8940f116ee1d6b94a2f90b13b5ea8d019b98bc8bafdcabcdd9bdbed" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/12/b3/231ffd4ab1fc9d679809f356cebee130ac7daa00d6d6f3206dd4fd137e9e/distro-1.9.0-py3-none-any.whl", hash = "sha256:7bffd925d65168f85027d8da9af6bddab658135b840670a223589bc0c8ef02b2" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
[[package]]
name = "docstring-parser"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "0.17.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/b2/9d/c3b43da9515bd270df0f80548d9944e389870713cc1fe2b8fb35fe2bcefd/docstring_parser-0.17.0.tar.gz", hash = "sha256:583de4a309722b3315439bb31d64ba3eebada841f2e2cee23b99df001434c912" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/55/e2/2537ebcff11c1ee1ff17d8d0b6f4db75873e3b0fb32c2d4a2ee31ecb310a/docstring_parser-0.17.0-py3-none-any.whl", hash = "sha256:cf2569abd23dce8099b300f9b4fa8191e9582dda731fd533daf54c4551658708" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "docutils"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "0.21.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ae/ed/aefcc8cd0ba62a0560c3c18c33925362d46c6075480bfa4df87b28e169a9/docutils-0.21.2.tar.gz", hash = "sha256:3a6b18732edf182daa3cd12775bbb338cf5691468f91eeeb109deff6ebfa986f" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/8f/d7/9322c609343d929e75e7e5e6255e614fcc67572cfd083959cdef3b7aad79/docutils-0.21.2-py3-none-any.whl", hash = "sha256:dafca5b9e384f0e419294eb4d2ff9fa826435bf15f15b7bd45723e8ad76811b2" },
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
]
[[package]]
name = "dropbox"
version = "12.0.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
dependencies = [
{ name = "requests" },
{ name = "six" },
{ name = "stone" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/9e/56/ac085f58e8e0d0bcafdf98c2605e454ac946e3d0c72679669ae112dc30be/dropbox-12.0.2.tar.gz", hash = "sha256:50057fd5ad5fcf047f542dfc6747a896e7ef982f1b5f8500daf51f3abd609962" }
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/2d/de/95d8204d9a20fbdb353c5f8e4229b0fcb90f22b96f8246ff1f47c8a45fd5/dropbox-12.0.2-py3-none-any.whl", hash = "sha256:c5b7e9c2668adb6b12dcecd84342565dc50f7d35ab6a748d155cb79040979d1c" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "duckduckgo-search"
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
version = "7.5.5"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "click" },
{ name = "lxml" },
{ name = "primp" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/13/dc/919d3d51ed702890a3e6e736e1e152d5d90856393200306e82fb54fde39e/duckduckgo_search-7.5.5.tar.gz", hash = "sha256:44ef03bfa5484bada786590f2d4c213251131765721383a177a0da6fa5c5e41a" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/fc/da/8376678b4a9ae0f9418d93df9c9cf851dced49c95ceb38daac6651e38f7a/duckduckgo_search-7.5.5-py3-none-any.whl", hash = "sha256:c71a0661aa436f215d9a05d653af424affb58825ab3e79f3b788053cbdee9ebc" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
[[package]]
name = "easygui"
version = "0.98.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/cc/ad/e35f7a30272d322be09dc98592d2f55d27cc933a7fde8baccbbeb2bd9409/easygui-0.98.3.tar.gz", hash = "sha256:d653ff79ee1f42f63b5a090f2f98ce02335d86ad8963b3ce2661805cafe99a04" }
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/8e/a7/b276ff776533b423710a285c8168b52551cb2ab0855443131fdc7fd8c16f/easygui-0.98.3-py2.py3-none-any.whl", hash = "sha256:33498710c68b5376b459cd3fc48d1d1f33822139eb3ed01defbc0528326da3ba" },
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
]
[[package]]
name = "ebcdic"
version = "1.1.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/0d/2f/633031205333bee5f9f93761af8268746aa75f38754823aabb8570eb245b/ebcdic-1.1.1-py2.py3-none-any.whl", hash = "sha256:33b4cb729bc2d0bf46cc1847b0e5946897cb8d3f53520c5b9aa5fa98d7e735f1" },
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "editdistance"
version = "0.8.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/d5/18/9f4f975ca87a390832b1c22478f3702fcdf739f83211e24d054b7551270d/editdistance-0.8.1.tar.gz", hash = "sha256:d1cdf80a5d5014b0c9126a69a42ce55a457b457f6986ff69ca98e4fe4d2d8fed" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "elastic-transport"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "8.17.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "certifi" },
{ name = "urllib3" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/6a/54/d498a766ac8fa475f931da85a154666cc81a70f8eb4a780bc8e4e934e9ac/elastic_transport-8.17.1.tar.gz", hash = "sha256:5edef32ac864dca8e2f0a613ef63491ee8d6b8cfb52881fa7313ba9290cac6d2" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/cf/cd/b71d5bc74cde7fc6fd9b2ff9389890f45d9762cbbbf81dc5e51fd7588c4a/elastic_transport-8.17.1-py3-none-any.whl", hash = "sha256:192718f498f1d10c5e9aa8b9cf32aed405e469a7f0e9d6a8923431dbb2c59fb8" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "elasticsearch"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "8.19.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "elastic-transport" },
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
{ name = "python-dateutil" },
{ name = "typing-extensions" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/6b/79/365e306017a9fcfbbefab1a3b588d2404bea8806b36766ff0f886509a20e/elasticsearch-8.19.3.tar.gz", hash = "sha256:e84dd618a220cac25b962790085045dd27ac72e01c0a5d81bd29a2d47a71f03f" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/56/0f/ac126833c385b06166d41c486e4911f58ad7791fd1a53dd6e0b8d16ff214/elasticsearch-8.19.3-py3-none-any.whl", hash = "sha256:fe1db2555811192e8a1be78b01234d0a49d32b185ea7eeeb6f059331dee32838" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "elasticsearch-dsl"
version = "8.12.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "elasticsearch" },
{ name = "python-dateutil" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/96/85/152eb3ed7af5f4d4a6cca563125491b61109a265a6e7a950a239209f4564/elasticsearch-dsl-8.12.0.tar.gz", hash = "sha256:ce32b8529888a97be911531e7590816cf3b1f608263eff6fb75aa7106e233c88" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/6b/ee/4699000ef357e476a3984fd1eff236f820e3346c4aef7c7772e580b81b31/elasticsearch_dsl-8.12.0-py3-none-any.whl", hash = "sha256:2ea9e6ded64d21a8f1ef72477a4d116c6fbeea631ac32a2e2490b9c0d09a99a6" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
[[package]]
name = "en-core-web-sm"
version = "3.8.0"
source = { url = "https://github.com/explosion/spacy-models/releases/download/en_core_web_sm-3.8.0/en_core_web_sm-3.8.0-py3-none-any.whl" }
wheels = [
{ url = "https://github.com/explosion/spacy-models/releases/download/en_core_web_sm-3.8.0/en_core_web_sm-3.8.0-py3-none-any.whl", hash = "sha256:1932429db727d4bff3deed6b34cfc05df17794f4a52eeb26cf8928f7c1a0fb85" },
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "et-xmlfile"
version = "2.0.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/d3/38/af70d7ab1ae9d4da450eeec1fa3918940a5fafb9055e934af8d6eb0c2313/et_xmlfile-2.0.0.tar.gz", hash = "sha256:dab3f4764309081ce75662649be815c4c9081e88f0837825f90fd28317d4da54" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/c1/8b/5fe2cc11fee489817272089c4203e679c63b570a5aaeb18d852ae3cbba6a/et_xmlfile-2.0.0-py3-none-any.whl", hash = "sha256:7a91720bc756843502c3b7504c77b8fe44217c85c537d85037f0f536151b2caa" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: Adds OpenSearch2.19.1 as the vector_database support (#7140) ### What problem does this PR solve? This PR adds the support for latest OpenSearch2.19.1 as the store engine & search engine option for RAGFlow. ### Main Benefit 1. OpenSearch2.19.1 is licensed under the [Apache v2.0 License] which is much better than Elasticsearch 2. For search, OpenSearch2.19.1 supports full-text search、vector_search、hybrid_search those are similar with Elasticsearch on schema 3. For store, OpenSearch2.19.1 stores text、vector those are quite simliar with Elasticsearch on schema ### Changes - Support opensearch_python_connetor. I make a lot of adaptions since the schema and api/method between ES and Opensearch differs in many ways(especially the knn_search has a significant gap) : rag/utils/opensearch_coon.py - Support static config adaptions by changing: conf/service_conf.yaml、api/settings.py、rag/settings.py - Supprt some store&search schema changes between OpenSearch and ES: conf/os_mapping.json - Support OpenSearch python sdk : pyproject.toml - Support docker config for OpenSearch2.19.1 : docker/.env、docker/docker-compose-base.yml、docker/service_conf.yaml.template ### How to use - I didn't change the priority that ES as the default doc/search engine. Only if in docker/.env , we set DOC_ENGINE=${DOC_ENGINE:-opensearch}, it will work. ### Others Our team tested a lot of docs in our environment by using OpenSearch as the vector database ,it works very well. All the conifg for OpenSearch is necessary. ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Yongteng Lei <yongtengrey@outlook.com> Co-authored-by: writinwaters <93570324+writinwaters@users.noreply.github.com> Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
2025-04-24 16:03:31 +08:00
[[package]]
name = "events"
version = "0.5"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Adds OpenSearch2.19.1 as the vector_database support (#7140) ### What problem does this PR solve? This PR adds the support for latest OpenSearch2.19.1 as the store engine & search engine option for RAGFlow. ### Main Benefit 1. OpenSearch2.19.1 is licensed under the [Apache v2.0 License] which is much better than Elasticsearch 2. For search, OpenSearch2.19.1 supports full-text search、vector_search、hybrid_search those are similar with Elasticsearch on schema 3. For store, OpenSearch2.19.1 stores text、vector those are quite simliar with Elasticsearch on schema ### Changes - Support opensearch_python_connetor. I make a lot of adaptions since the schema and api/method between ES and Opensearch differs in many ways(especially the knn_search has a significant gap) : rag/utils/opensearch_coon.py - Support static config adaptions by changing: conf/service_conf.yaml、api/settings.py、rag/settings.py - Supprt some store&search schema changes between OpenSearch and ES: conf/os_mapping.json - Support OpenSearch python sdk : pyproject.toml - Support docker config for OpenSearch2.19.1 : docker/.env、docker/docker-compose-base.yml、docker/service_conf.yaml.template ### How to use - I didn't change the priority that ES as the default doc/search engine. Only if in docker/.env , we set DOC_ENGINE=${DOC_ENGINE:-opensearch}, it will work. ### Others Our team tested a lot of docs in our environment by using OpenSearch as the vector database ,it works very well. All the conifg for OpenSearch is necessary. ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Yongteng Lei <yongtengrey@outlook.com> Co-authored-by: writinwaters <93570324+writinwaters@users.noreply.github.com> Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
2025-04-24 16:03:31 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/25/ed/e47dec0626edd468c84c04d97769e7ab4ea6457b7f54dcb3f72b17fcd876/Events-0.5-py3-none-any.whl", hash = "sha256:a7286af378ba3e46640ac9825156c93bdba7502174dd696090fdfcd4d80a1abd" },
Feat: Adds OpenSearch2.19.1 as the vector_database support (#7140) ### What problem does this PR solve? This PR adds the support for latest OpenSearch2.19.1 as the store engine & search engine option for RAGFlow. ### Main Benefit 1. OpenSearch2.19.1 is licensed under the [Apache v2.0 License] which is much better than Elasticsearch 2. For search, OpenSearch2.19.1 supports full-text search、vector_search、hybrid_search those are similar with Elasticsearch on schema 3. For store, OpenSearch2.19.1 stores text、vector those are quite simliar with Elasticsearch on schema ### Changes - Support opensearch_python_connetor. I make a lot of adaptions since the schema and api/method between ES and Opensearch differs in many ways(especially the knn_search has a significant gap) : rag/utils/opensearch_coon.py - Support static config adaptions by changing: conf/service_conf.yaml、api/settings.py、rag/settings.py - Supprt some store&search schema changes between OpenSearch and ES: conf/os_mapping.json - Support OpenSearch python sdk : pyproject.toml - Support docker config for OpenSearch2.19.1 : docker/.env、docker/docker-compose-base.yml、docker/service_conf.yaml.template ### How to use - I didn't change the priority that ES as the default doc/search engine. Only if in docker/.env , we set DOC_ENGINE=${DOC_ENGINE:-opensearch}, it will work. ### Others Our team tested a lot of docs in our environment by using OpenSearch as the vector database ,it works very well. All the conifg for OpenSearch is necessary. ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Yongteng Lei <yongtengrey@outlook.com> Co-authored-by: writinwaters <93570324+writinwaters@users.noreply.github.com> Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
2025-04-24 16:03:31 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "exceptiongroup"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "1.3.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/50/79/66800aadf48771f6b62f7eb014e352e5d06856655206165d775e675a02c9/exceptiongroup-1.3.1.tar.gz", hash = "sha256:8b412432c6055b0b7d14c310000ae93352ed6754f70fa8f7c34141f91c4e3219" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/8a/0e/97c33bf5009bdbac74fd2beace167cab3f978feb69cc36f1ef79360d6c4e/exceptiongroup-1.3.1-py3-none-any.whl", hash = "sha256:a7a39a3bd276781e98394987d3a5701d0c4edffb633bb7a5144577f82c773598" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Add ut test into CI (#11753) ### What problem does this PR solve? As title ### Type of change - [x] Other (please describe): --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-05 11:40:16 +08:00
[[package]]
name = "execnet"
version = "2.1.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/bf/89/780e11f9588d9e7128a3f87788354c7946a9cbb1401ad38a48c4db9a4f07/execnet-2.1.2.tar.gz", hash = "sha256:63d83bfdd9a23e35b9c6a3261412324f964c2ec8dcd8d3c6916ee9373e0befcd" }
Add ut test into CI (#11753) ### What problem does this PR solve? As title ### Type of change - [x] Other (please describe): --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-05 11:40:16 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ab/84/02fc1827e8cdded4aa65baef11296a9bbe595c474f0d6d758af082d849fd/execnet-2.1.2-py3-none-any.whl", hash = "sha256:67fba928dd5a544b783f6056f449e5e3931a5c378b128bc18501f7ea79e296ec" },
Add ut test into CI (#11753) ### What problem does this PR solve? As title ### Type of change - [x] Other (please describe): --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-05 11:40:16 +08:00
]
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
[[package]]
name = "extract-msg"
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
version = "0.55.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
dependencies = [
{ name = "beautifulsoup4" },
{ name = "compressed-rtf" },
{ name = "ebcdic" },
{ name = "olefile" },
{ name = "red-black-tree-mod" },
{ name = "rtfde" },
{ name = "tzlocal" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/5e/65/c70afb3b119a44b3ee36b029485dc15326cf3a7c50da19a1ecbbf949c5d1/extract_msg-0.55.0.tar.gz", hash = "sha256:cf08283498c3dfcc7f894dad1579f52e3ced9fb76b865c2355cbe757af8a54e1" }
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/53/81/87d5241036046ea17c5c8db228f4c9e04e07e53b627015d4496a99449aaf/extract_msg-0.55.0-py3-none-any.whl", hash = "sha256:baf0cdee9a8d267b70c366bc57ceb03dbfa1e7ab2dca6824169a7fe623f0917c" },
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "fake-useragent"
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
version = "2.2.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/41/43/948d10bf42735709edb5ae51e23297d034086f17fc7279fef385a7acb473/fake_useragent-2.2.0.tar.gz", hash = "sha256:4e6ab6571e40cc086d788523cf9e018f618d07f9050f822ff409a4dfe17c16b2" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/51/37/b3ea9cd5558ff4cb51957caca2193981c6b0ff30bd0d2630ac62505d99d0/fake_useragent-2.2.0-py3-none-any.whl", hash = "sha256:67f35ca4d847b0d298187443aaf020413746e56acd985a611908c73dba2daa24" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "fastavro"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "1.12.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/65/8b/fa2d3287fd2267be6261d0177c6809a7fa12c5600ddb33490c8dc29e77b2/fastavro-1.12.1.tar.gz", hash = "sha256:2f285be49e45bc047ab2f6bed040bb349da85db3f3c87880e4b92595ea093b2b" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/bb/57/26d5efef9182392d5ac9f253953c856ccb66e4c549fd3176a1e94efb05c9/fastavro-1.12.1-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:78df838351e4dff9edd10a1c41d1324131ffecbadefb9c297d612ef5363c049a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/33/cb/8ab55b21d018178eb126007a56bde14fd01c0afc11d20b5f2624fe01e698/fastavro-1.12.1-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:780476c23175d2ae457c52f45b9ffa9d504593499a36cd3c1929662bf5b7b14b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fe/03/9c94ec9bf873eb1ffb0aa694f4e71940154e6e9728ddfdc46046d7e8ced4/fastavro-1.12.1-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:0714b285160fcd515eb0455540f40dd6dac93bdeacdb03f24e8eac3d8aa51f8d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/75/c8/cb472347c5a584ccb8777a649ebb28278fccea39d005fc7df19996f41df8/fastavro-1.12.1-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:a8bc2dcec5843d499f2489bfe0747999108f78c5b29295d877379f1972a3d41a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e1/77/569ce9474c40304b3a09e109494e020462b83e405545b78069ddba5f614e/fastavro-1.12.1-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:3b1921ac35f3d89090a5816b626cf46e67dbecf3f054131f84d56b4e70496f45" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4a/1f/9589e35e9ea68035385db7bdbf500d36b8891db474063fb1ccc8215ee37c/fastavro-1.12.1-cp313-cp313-win_amd64.whl", hash = "sha256:5aa777b8ee595b50aa084104cd70670bf25a7bbb9fd8bb5d07524b0785ee1699" },
{ url = "https://mirrors.aliyun.com/pypi/packages/6c/d2/78435fe737df94bd8db2234b2100f5453737cffd29adee2504a2b013de84/fastavro-1.12.1-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:c3d67c47f177e486640404a56f2f50b165fe892cc343ac3a34673b80cc7f1dd6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b6/be/428f99b10157230ddac77ec8cc167005b29e2bd5cbe228345192bb645f30/fastavro-1.12.1-cp313-cp313t-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:5217f773492bac43dae15ff2931432bce2d7a80be7039685a78d3fab7df910bd" },
{ url = "https://mirrors.aliyun.com/pypi/packages/16/08/a2eea4f20b85897740efe44887e1ac08f30dfa4bfc3de8962bdcbb21a5a1/fastavro-1.12.1-cp313-cp313t-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:469fecb25cba07f2e1bfa4c8d008477cd6b5b34a59d48715e1b1a73f6160097d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/87/bb/b4c620b9eb6e9838c7f7e4b7be0762834443adf9daeb252a214e9ad3178c/fastavro-1.12.1-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:d71c8aa841ef65cfab709a22bb887955f42934bced3ddb571e98fdbdade4c609" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3d/d1/e69534ccdd5368350646fea7d93be39e5f77c614cca825c990bd9ca58f67/fastavro-1.12.1-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:b81fc04e85dfccf7c028e0580c606e33aa8472370b767ef058aae2c674a90746" },
Removed onnxruntime (#4632) ### What problem does this PR solve? Removed onnxruntime. It conflicts with the onnxruntime-gpu. ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-01-24 23:41:52 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "fastparquet"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "2025.12.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "cramjam" },
{ name = "fsspec" },
{ name = "numpy" },
{ name = "packaging" },
{ name = "pandas" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/1e/ad/87f7f5750685e8e0a359d732c85332481ba9b5723af579f8755f81154d0b/fastparquet-2025.12.0.tar.gz", hash = "sha256:85f807d3846c7691855a68ed7ff6ee40654b72b997f5b1199e6310a1e19d1cd5" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/13/cf/1801afbc1e84ad0413ec66bf93590472152462c454593e3be3265861aa0f/fastparquet-2025.12.0-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:1bd79ca75977aaeaae8d2a6cb1958e806991f0ff23207b938522a59a724491b2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/79/f9/5539b19ae7e1e0ad77f5b8a1e8d480fdf0193639cf97239734173b8730ab/fastparquet-2025.12.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:b6db801b72433d8227fcb92009a631f14d6d49a43b3c599911b58a8a6ffde9e3" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ff/d9/0f39782c500bbf6b2e40a67cac3c9ec2eae70bdaa8b283106c2b3d532a95/fastparquet-2025.12.0-cp313-cp313-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:23cce7202de91b64abb251cec07125d94e8108eb99aab6ffa42570a89a5c869d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b5/16/d0d0c5ca6a9fa13e2f36e6983452d798d8116bd5d05bf23246efd1c23dc8/fastparquet-2025.12.0-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:038c3ed1f211f538cd03df7b053cc842677efd5832e37b000a8c721584ff42b4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/eb/26/6c6a1cae46104a3ec5da87cb5fefb3eac0c07f04e56786f928164942e91a/fastparquet-2025.12.0-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:424ffcfc89c678eb8e695ff882d114e46beda8b7e13be58b6793f2ee07c84a6f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/69/77/6a7158e2817d44fb80f32a4a4c3f8cadf7e273fac34e04155588bf2b3141/fastparquet-2025.12.0-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:f25aae3e585dd033ed02ee167a825bf1fcb440629c63f7d59d6c4d2789c327a3" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ee/89/58b1d885dcf05ba619d3a9bbf61b3bff611c4636880077be8659bf29ce94/fastparquet-2025.12.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:90ac4a51e5acb2644ec111532c8fcfc128efcc351ba2ee914394a58460310b93" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f8/10/380cba3ee18b25384cbf0d229b8cad47d63eb89c630f267cf1e11c64fe16/fastparquet-2025.12.0-cp313-cp313-win_amd64.whl", hash = "sha256:7ac92db3b3200fe3be07363277678bfd532c6723510b40c20510631ca434a049" },
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
]
[[package]]
name = "fastuuid"
version = "0.14.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/c3/7d/d9daedf0f2ebcacd20d599928f8913e9d2aea1d56d2d355a93bfa2b611d7/fastuuid-0.14.0.tar.gz", hash = "sha256:178947fc2f995b38497a74172adee64fdeb8b7ec18f2a5934d037641ba265d26" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/a5/83/ae12dd39b9a39b55d7f90abb8971f1a5f3c321fd72d5aa83f90dc67fe9ed/fastuuid-0.14.0-cp313-cp313-macosx_10_12_x86_64.macosx_11_0_arm64.macosx_10_12_universal2.whl", hash = "sha256:77a09cb7427e7af74c594e409f7731a0cf887221de2f698e1ca0ebf0f3139021" },
{ url = "https://mirrors.aliyun.com/pypi/packages/53/b0/a4b03ff5d00f563cc7546b933c28cb3f2a07344b2aec5834e874f7d44143/fastuuid-0.14.0-cp313-cp313-macosx_10_12_x86_64.whl", hash = "sha256:9bd57289daf7b153bfa3e8013446aa144ce5e8c825e9e366d455155ede5ea2dc" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9c/6d/64aee0a0f6a58eeabadd582e55d0d7d70258ffdd01d093b30c53d668303b/fastuuid-0.14.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:ac60fc860cdf3c3f327374db87ab8e064c86566ca8c49d2e30df15eda1b0c2d5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/60/f5/a7e9cda8369e4f7919d36552db9b2ae21db7915083bc6336f1b0082c8b2e/fastuuid-0.14.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:ab32f74bd56565b186f036e33129da77db8be09178cd2f5206a5d4035fb2a23f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f0/d3/8ce11827c783affffd5bd4d6378b28eb6cc6d2ddf41474006b8d62e7448e/fastuuid-0.14.0-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:33e678459cf4addaedd9936bbb038e35b3f6b2061330fd8f2f6a1d80414c0f87" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a2/51/680fb6352d0bbade04036da46264a8001f74b7484e2fd1f4da9e3db1c666/fastuuid-0.14.0-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:1e3cc56742f76cd25ecb98e4b82a25f978ccffba02e4bdce8aba857b6d85d87b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fa/7c/2014b5785bd8ebdab04ec857635ebd84d5ee4950186a577db9eff0fb8ff6/fastuuid-0.14.0-cp313-cp313-musllinux_1_1_aarch64.whl", hash = "sha256:cb9a030f609194b679e1660f7e32733b7a0f332d519c5d5a6a0a580991290022" },
{ url = "https://mirrors.aliyun.com/pypi/packages/01/d2/524d4ceeba9160e7a9bc2ea3e8f4ccf1ad78f3bde34090ca0c51f09a5e91/fastuuid-0.14.0-cp313-cp313-musllinux_1_1_i686.whl", hash = "sha256:09098762aad4f8da3a888eb9ae01c84430c907a297b97166b8abc07b640f2995" },
{ url = "https://mirrors.aliyun.com/pypi/packages/bc/17/354d04951ce114bf4afc78e27a18cfbd6ee319ab1829c2d5fb5e94063ac6/fastuuid-0.14.0-cp313-cp313-musllinux_1_1_x86_64.whl", hash = "sha256:1383fff584fa249b16329a059c68ad45d030d5a4b70fb7c73a08d98fd53bcdab" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fb/be/d7be8670151d16d88f15bb121c5b66cdb5ea6a0c2a362d0dcf30276ade53/fastuuid-0.14.0-cp313-cp313-win32.whl", hash = "sha256:a0809f8cc5731c066c909047f9a314d5f536c871a7a22e815cc4967c110ac9ad" },
{ url = "https://mirrors.aliyun.com/pypi/packages/22/1d/5573ef3624ceb7abf4a46073d3554e37191c868abc3aecd5289a72f9810a/fastuuid-0.14.0-cp313-cp313-win_amd64.whl", hash = "sha256:0df14e92e7ad3276327631c9e7cec09e32572ce82089c55cb1bb8df71cf394ed" },
Don't release full image (#10654) ### What problem does this PR solve? Introduced gpu profile in .env Added Dockerfile_tei fix datrie Removed LIGHTEN flag ### Type of change - [x] Documentation Update - [x] Refactoring
2025-10-23 23:02:27 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "feedparser"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "6.0.12"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "sgmllib3k" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/dc/79/db7edb5e77d6dfbc54d7d9df72828be4318275b2e580549ff45a962f6461/feedparser-6.0.12.tar.gz", hash = "sha256:64f76ce90ae3e8ef5d1ede0f8d3b50ce26bcce71dd8ae5e82b1cd2d4a5f94228" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/4e/eb/c96d64137e29ae17d83ad2552470bafe3a7a915e85434d9942077d7fd011/feedparser-6.0.12-py3-none-any.whl", hash = "sha256:6bbff10f5a52662c00a2e3f86a38928c37c48f77b3c511aedcd51de933549324" },
Feat:new api /sequence2txt and update QWenSeq2txt (#11643) ### What problem does this PR solve? change: new api /sequence2txt, update QWenSeq2txt and ZhipuSeq2txt ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-02 11:17:31 +08:00
]
[[package]]
name = "ffmpeg-python"
version = "0.2.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat:new api /sequence2txt and update QWenSeq2txt (#11643) ### What problem does this PR solve? change: new api /sequence2txt, update QWenSeq2txt and ZhipuSeq2txt ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-02 11:17:31 +08:00
dependencies = [
{ name = "future" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/dd/5e/d5f9105d59c1325759d838af4e973695081fbbc97182baf73afc78dec266/ffmpeg-python-0.2.0.tar.gz", hash = "sha256:65225db34627c578ef0e11c8b1eb528bb35e024752f6f10b78c011f6f64c4127" }
Feat:new api /sequence2txt and update QWenSeq2txt (#11643) ### What problem does this PR solve? change: new api /sequence2txt, update QWenSeq2txt and ZhipuSeq2txt ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-02 11:17:31 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/d7/0c/56be52741f75bad4dc6555991fabd2e07b432d333da82c11ad701123888a/ffmpeg_python-0.2.0-py3-none-any.whl", hash = "sha256:ac441a0404e053f8b6a1113a77c0f452f1cfc62f6344a769475ffdc0f56c23c5" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "filelock"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "3.20.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/1d/65/ce7f1b70157833bf3cb851b556a37d4547ceafc158aa9b34b36782f23696/filelock-3.20.3.tar.gz", hash = "sha256:18c57ee915c7ec61cff0ecf7f0f869936c7c30191bb0cf406f1341778d0834e1" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/b5/36/7fb70f04bf00bc646cd5bb45aa9eddb15e19437a28b8fb2b4a5249fac770/filelock-3.20.3-py3-none-any.whl", hash = "sha256:4b0dda527ee31078689fc205ec4f1c1bf7d56cf88b6dc9426c4f230e46c2dce1" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "flasgger"
version = "0.9.7.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "flask" },
{ name = "jsonschema" },
{ name = "mistune" },
{ name = "packaging" },
{ name = "pyyaml" },
{ name = "six" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/8a/e4/05e80adeadc39f171b51bd29b24a6d9838127f3aaa1b07c1501e662a8cee/flasgger-0.9.7.1.tar.gz", hash = "sha256:ca098e10bfbb12f047acc6299cc70a33851943a746e550d86e65e60d4df245fb" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "flask"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "3.1.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "blinker" },
{ name = "click" },
{ name = "itsdangerous" },
{ name = "jinja2" },
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
{ name = "markupsafe" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "werkzeug" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/dc/6d/cfe3c0fcc5e477df242b98bfe186a4c34357b4847e87ecaef04507332dab/flask-3.1.2.tar.gz", hash = "sha256:bf656c15c80190ed628ad08cdfd3aaa35beb087855e2f494910aa3774cc4fd87" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ec/f9/7f9263c5695f4bd0023734af91bedb2ff8209e8de6ead162f35d8dc762fd/flask-3.1.2-py3-none-any.whl", hash = "sha256:ca1d8112ec8a6158cc29ea4858963350011b5c846a414cdb7a954aa9e967d03c" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "flask-cors"
Fix security issue (#11965) ### What problem does this PR solve? - CVE-2024-6866 - CVE-2024-6844 - CVE-2024-6839 ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue) Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-16 13:31:45 +08:00
version = "6.0.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "flask" },
Fix security issue (#11965) ### What problem does this PR solve? - CVE-2024-6866 - CVE-2024-6844 - CVE-2024-6839 ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue) Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-16 13:31:45 +08:00
{ name = "werkzeug" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/70/74/0fc0fa68d62f21daef41017dafab19ef4b36551521260987eb3a5394c7ba/flask_cors-6.0.2.tar.gz", hash = "sha256:6e118f3698249ae33e429760db98ce032a8bf9913638d085ca0f4c5534ad2423" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/4f/af/72ad54402e599152de6d067324c46fe6a4f531c7c65baf7e96c63db55eaf/flask_cors-6.0.2-py3-none-any.whl", hash = "sha256:e57544d415dfd7da89a9564e1e3a9e515042df76e12130641ca6f3f2f03b699a" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "flask-login"
version = "0.6.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "flask" },
{ name = "werkzeug" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/c3/6e/2f4e13e373bb49e68c02c51ceadd22d172715a06716f9299d9df01b6ddb2/Flask-Login-0.6.3.tar.gz", hash = "sha256:5e23d14a607ef12806c699590b89d0f0e0d67baeec599d75947bf9c147330333" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/59/f5/67e9cc5c2036f58115f9fe0f00d203cf6780c3ff8ae0e705e7a9d9e8ff9e/Flask_Login-0.6.3-py3-none-any.whl", hash = "sha256:849b25b82a436bf830a054e74214074af59097171562ab10bfa999e6b78aae5d" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: add SMTP support for user invitation emails (#9479) ### What problem does this PR solve? Add SMTP support for user invitation emails ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-08-15 18:12:20 +08:00
[[package]]
name = "flask-mail"
version = "0.10.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: add SMTP support for user invitation emails (#9479) ### What problem does this PR solve? Add SMTP support for user invitation emails ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-08-15 18:12:20 +08:00
dependencies = [
{ name = "blinker" },
{ name = "flask" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ba/29/e92dc84c675d1e8d260d5768eb3fb65c70cbd33addecf424187587bee862/flask_mail-0.10.0.tar.gz", hash = "sha256:44083e7b02bbcce792209c06252f8569dd5a325a7aaa76afe7330422bd97881d" }
Feat: add SMTP support for user invitation emails (#9479) ### What problem does this PR solve? Add SMTP support for user invitation emails ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-08-15 18:12:20 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/e4/c0/a81083da779f482494d49195d8b6c9fde21072558253e4a9fb2ec969c3c1/flask_mail-0.10.0-py3-none-any.whl", hash = "sha256:a451e490931bb3441d9b11ebab6812a16bfa81855792ae1bf9c1e1e22c4e51e7" },
Feat: add SMTP support for user invitation emails (#9479) ### What problem does this PR solve? Add SMTP support for user invitation emails ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-08-15 18:12:20 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "flask-session"
version = "0.8.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "cachelib" },
{ name = "flask" },
{ name = "msgspec" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/86/d7/0ba4180513abe28eadc208123c76f9f09e290d5939fb2eb68323b9733354/flask_session-0.8.0.tar.gz", hash = "sha256:20e045eb01103694e70be4a49f3a80dbb1b57296a22dc6f44bbf3f83ef0742ff" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/67/1b/f085ceebb825d1cfaf078852b67cd248a33af2905f40ba9860cc006d966b/flask_session-0.8.0-py3-none-any.whl", hash = "sha256:5dae6e9ddab334f8dc4dea4305af37851f4e7dc0f484caf3351184001195e3b7" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "flatbuffers"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "25.12.19"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/e8/2d/d2a548598be01649e2d46231d151a6c56d10b964d94043a335ae56ea2d92/flatbuffers-25.12.19-py2.py3-none-any.whl", hash = "sha256:7634f50c427838bb021c2d66a3d1168e9d199b0607e6329399f04846d42e20b4" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "fonttools"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "4.61.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ec/ca/cf17b88a8df95691275a3d77dc0a5ad9907f328ae53acbe6795da1b2f5ed/fonttools-4.61.1.tar.gz", hash = "sha256:6675329885c44657f826ef01d9e4fb33b9158e9d93c537d84ad8399539bc6f69" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/4b/cf/00ba28b0990982530addb8dc3e9e6f2fa9cb5c20df2abdda7baa755e8fe1/fonttools-4.61.1-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:8c56c488ab471628ff3bfa80964372fc13504ece601e0d97a78ee74126b2045c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/5a/ca/468c9a8446a2103ae645d14fee3f610567b7042aba85031c1c65e3ef7471/fonttools-4.61.1-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:dc492779501fa723b04d0ab1f5be046797fee17d27700476edc7ee9ae535a61e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a3/4b/d67eedaed19def5967fade3297fed8161b25ba94699efc124b14fb68cdbc/fonttools-4.61.1-cp313-cp313-manylinux1_x86_64.manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:64102ca87e84261419c3747a0d20f396eb024bdbeb04c2bfb37e2891f5fadcb5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b0/8d/6fb3494dfe61a46258cd93d979cf4725ded4eb46c2a4ca35e4490d84daea/fonttools-4.61.1-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:4c1b526c8d3f615a7b1867f38a9410849c8f4aef078535742198e942fba0e9bd" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f7/f1/a47f1d30b3dc00d75e7af762652d4cbc3dff5c2697a0dbd5203c81afd9c3/fonttools-4.61.1-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:41ed4b5ec103bd306bb68f81dc166e77409e5209443e5773cb4ed837bcc9b0d3" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a7/01/e6ae64a0981076e8a66906fab01539799546181e32a37a0257b77e4aa88b/fonttools-4.61.1-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:b501c862d4901792adaec7c25b1ecc749e2662543f68bb194c42ba18d6eec98d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/73/aa/28e40b8d6809a9b5075350a86779163f074d2b617c15d22343fce81918db/fonttools-4.61.1-cp313-cp313-win32.whl", hash = "sha256:4d7092bb38c53bbc78e9255a59158b150bcdc115a1e3b3ce0b5f267dc35dd63c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1a/59/453c06d1d83dc0951b69ef692d6b9f1846680342927df54e9a1ca91c6f90/fonttools-4.61.1-cp313-cp313-win_amd64.whl", hash = "sha256:21e7c8d76f62ab13c9472ccf74515ca5b9a761d1bde3265152a6dc58700d895b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c7/4e/ce75a57ff3aebf6fc1f4e9d508b8e5810618a33d900ad6c19eb30b290b97/fonttools-4.61.1-py3-none-any.whl", hash = "sha256:17d2bf5d541add43822bcf0c43d7d847b160c9bb01d15d5007d84e2217aaa371" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "free-proxy"
version = "1.1.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "lxml" },
{ name = "requests" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/d5/10/3654b44093aa3e587948c770279baca3a8dfe4d14a616142e8c6bf04b09b/free_proxy-1.1.3.tar.gz", hash = "sha256:6d82aa112e3df7725bdbf177e2110bccdf5f3bbd6e1c70b8616ec12ae3bbf98c" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
[[package]]
name = "frozendict"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "2.4.7"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/90/b2/2a3d1374b7780999d3184e171e25439a8358c47b481f68be883c14086b4c/frozendict-2.4.7.tar.gz", hash = "sha256:e478fb2a1391a56c8a6e10cc97c4a9002b410ecd1ac28c18d780661762e271bd" }
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/38/74/f94141b38a51a553efef7f510fc213894161ae49b88bffd037f8d2a7cb2f/frozendict-2.4.7-py3-none-any.whl", hash = "sha256:972af65924ea25cf5b4d9326d549e69a9a4918d8a76a9d3a7cd174d98b237550" },
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "frozenlist"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "1.8.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/2d/f5/c831fac6cc817d26fd54c7eaccd04ef7e0288806943f7cc5bbf69f3ac1f0/frozenlist-1.8.0.tar.gz", hash = "sha256:3ede829ed8d842f6cd48fc7081d7a41001a56f1f38603f9d49bf3020d59a31ad" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/2d/40/0832c31a37d60f60ed79e9dfb5a92e1e2af4f40a16a29abcc7992af9edff/frozenlist-1.8.0-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:8d92f1a84bb12d9e56f818b3a746f3efba93c1b63c8387a73dde655e1e42282a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/30/ba/b0b3de23f40bc55a7057bd38434e25c34fa48e17f20ee273bbde5e0650f3/frozenlist-1.8.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:96153e77a591c8adc2ee805756c61f59fef4cf4073a9275ee86fe8cba41241f7" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0c/ab/6e5080ee374f875296c4243c381bbdef97a9ac39c6e3ce1d5f7d42cb78d6/frozenlist-1.8.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:f21f00a91358803399890ab167098c131ec2ddd5f8f5fd5fe9c9f2c6fcd91e40" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d5/4e/e4691508f9477ce67da2015d8c00acd751e6287739123113a9fca6f1604e/frozenlist-1.8.0-cp313-cp313-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:fb30f9626572a76dfe4293c7194a09fb1fe93ba94c7d4f720dfae3b646b45027" },
{ url = "https://mirrors.aliyun.com/pypi/packages/40/76/c202df58e3acdf12969a7895fd6f3bc016c642e6726aa63bd3025e0fc71c/frozenlist-1.8.0-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:eaa352d7047a31d87dafcacbabe89df0aa506abb5b1b85a2fb91bc3faa02d822" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f9/c0/8746afb90f17b73ca5979c7a3958116e105ff796e718575175319b5bb4ce/frozenlist-1.8.0-cp313-cp313-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:03ae967b4e297f58f8c774c7eabcce57fe3c2434817d4385c50661845a058121" },
{ url = "https://mirrors.aliyun.com/pypi/packages/7e/eb/4c7eefc718ff72f9b6c4893291abaae5fbc0c82226a32dcd8ef4f7a5dbef/frozenlist-1.8.0-cp313-cp313-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:f6292f1de555ffcc675941d65fffffb0a5bcd992905015f85d0592201793e0e5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c2/4e/e5c02187cf704224f8b21bee886f3d713ca379535f16893233b9d672ea71/frozenlist-1.8.0-cp313-cp313-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:29548f9b5b5e3460ce7378144c3010363d8035cea44bc0bf02d57f5a685e084e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1f/96/cb85ec608464472e82ad37a17f844889c36100eed57bea094518bf270692/frozenlist-1.8.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:ec3cc8c5d4084591b4237c0a272cc4f50a5b03396a47d9caaf76f5d7b38a4f11" },
{ url = "https://mirrors.aliyun.com/pypi/packages/5d/6f/4ae69c550e4cee66b57887daeebe006fe985917c01d0fff9caab9883f6d0/frozenlist-1.8.0-cp313-cp313-musllinux_1_2_armv7l.whl", hash = "sha256:517279f58009d0b1f2e7c1b130b377a349405da3f7621ed6bfae50b10adf20c1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/7a/58/afd56de246cf11780a40a2c28dc7cbabbf06337cc8ddb1c780a2d97e88d8/frozenlist-1.8.0-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:db1e72ede2d0d7ccb213f218df6a078a9c09a7de257c2fe8fcef16d5925230b1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/cb/36/cdfaf6ed42e2644740d4a10452d8e97fa1c062e2a8006e4b09f1b5fd7d63/frozenlist-1.8.0-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:b4dec9482a65c54a5044486847b8a66bf10c9cb4926d42927ec4e8fd5db7fed8" },
{ url = "https://mirrors.aliyun.com/pypi/packages/03/a8/9ea226fbefad669f11b52e864c55f0bd57d3c8d7eb07e9f2e9a0b39502e1/frozenlist-1.8.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:21900c48ae04d13d416f0e1e0c4d81f7931f73a9dfa0b7a8746fb2fe7dd970ed" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1e/0b/1b5531611e83ba7d13ccc9988967ea1b51186af64c42b7a7af465dcc9568/frozenlist-1.8.0-cp313-cp313-win32.whl", hash = "sha256:8b7b94a067d1c504ee0b16def57ad5738701e4ba10cec90529f13fa03c833496" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d8/cf/174c91dbc9cc49bc7b7aab74d8b734e974d1faa8f191c74af9b7e80848e6/frozenlist-1.8.0-cp313-cp313-win_amd64.whl", hash = "sha256:878be833caa6a3821caf85eb39c5ba92d28e85df26d57afb06b35b2efd937231" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c1/17/502cd212cbfa96eb1388614fe39a3fc9ab87dbbe042b66f97acb57474834/frozenlist-1.8.0-cp313-cp313-win_arm64.whl", hash = "sha256:44389d135b3ff43ba8cc89ff7f51f5a0bb6b63d829c8300f79a2fe4fe61bcc62" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d2/5c/3bbfaa920dfab09e76946a5d2833a7cbdf7b9b4a91c714666ac4855b88b4/frozenlist-1.8.0-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:e25ac20a2ef37e91c1b39938b591457666a0fa835c7783c3a8f33ea42870db94" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d2/d6/f03961ef72166cec1687e84e8925838442b615bd0b8854b54923ce5b7b8a/frozenlist-1.8.0-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:07cdca25a91a4386d2e76ad992916a85038a9b97561bf7a3fd12d5d9ce31870c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1e/bb/a6d12b7ba4c3337667d0e421f7181c82dda448ce4e7ad7ecd249a16fa806/frozenlist-1.8.0-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:4e0c11f2cc6717e0a741f84a527c52616140741cd812a50422f83dc31749fb52" },
{ url = "https://mirrors.aliyun.com/pypi/packages/bc/71/d1fed0ffe2c2ccd70b43714c6cab0f4188f09f8a67a7914a6b46ee30f274/frozenlist-1.8.0-cp313-cp313t-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:b3210649ee28062ea6099cfda39e147fa1bc039583c8ee4481cb7811e2448c51" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c9/1f/fb1685a7b009d89f9bf78a42d94461bc06581f6e718c39344754a5d9bada/frozenlist-1.8.0-cp313-cp313t-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:581ef5194c48035a7de2aefc72ac6539823bb71508189e5de01d60c9dcd5fa65" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e6/3b/b991fe1612703f7e0d05c0cf734c1b77aaf7c7d321df4572e8d36e7048c8/frozenlist-1.8.0-cp313-cp313t-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:3ef2d026f16a2b1866e1d86fc4e1291e1ed8a387b2c333809419a2f8b3a77b82" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ca/ec/c5c618767bcdf66e88945ec0157d7f6c4a1322f1473392319b7a2501ded7/frozenlist-1.8.0-cp313-cp313t-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:5500ef82073f599ac84d888e3a8c1f77ac831183244bfd7f11eaa0289fb30714" },
{ url = "https://mirrors.aliyun.com/pypi/packages/7c/ce/3934758637d8f8a88d11f0585d6495ef54b2044ed6ec84492a91fa3b27aa/frozenlist-1.8.0-cp313-cp313t-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:50066c3997d0091c411a66e710f4e11752251e6d2d73d70d8d5d4c76442a199d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fc/4f/a7e4d0d467298f42de4b41cbc7ddaf19d3cfeabaf9ff97c20c6c7ee409f9/frozenlist-1.8.0-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:5c1c8e78426e59b3f8005e9b19f6ff46e5845895adbde20ece9218319eca6506" },
{ url = "https://mirrors.aliyun.com/pypi/packages/dc/48/c7b163063d55a83772b268e6d1affb960771b0e203b632cfe09522d67ea5/frozenlist-1.8.0-cp313-cp313t-musllinux_1_2_armv7l.whl", hash = "sha256:eefdba20de0d938cec6a89bd4d70f346a03108a19b9df4248d3cf0d88f1b0f51" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9f/d0/2366d3c4ecdc2fd391e0afa6e11500bfba0ea772764d631bbf82f0136c9d/frozenlist-1.8.0-cp313-cp313t-musllinux_1_2_ppc64le.whl", hash = "sha256:cf253e0e1c3ceb4aaff6df637ce033ff6535fb8c70a764a8f46aafd3d6ab798e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b8/94/daff920e82c1b70e3618a2ac39fbc01ae3e2ff6124e80739ce5d71c9b920/frozenlist-1.8.0-cp313-cp313t-musllinux_1_2_s390x.whl", hash = "sha256:032efa2674356903cd0261c4317a561a6850f3ac864a63fc1583147fb05a79b0" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e3/20/bba307ab4235a09fdcd3cc5508dbabd17c4634a1af4b96e0f69bfe551ebd/frozenlist-1.8.0-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:6da155091429aeba16851ecb10a9104a108bcd32f6c1642867eadaee401c1c41" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fd/00/04ca1c3a7a124b6de4f8a9a17cc2fcad138b4608e7a3fc5877804b8715d7/frozenlist-1.8.0-cp313-cp313t-win32.whl", hash = "sha256:0f96534f8bfebc1a394209427d0f8a63d343c9779cda6fc25e8e121b5fd8555b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/59/5e/c69f733a86a94ab10f68e496dc6b7e8bc078ebb415281d5698313e3af3a1/frozenlist-1.8.0-cp313-cp313t-win_amd64.whl", hash = "sha256:5d63a068f978fc69421fb0e6eb91a9603187527c86b7cd3f534a5b77a592b888" },
{ url = "https://mirrors.aliyun.com/pypi/packages/16/6c/be9d79775d8abe79b05fa6d23da99ad6e7763a1d080fbae7290b286093fd/frozenlist-1.8.0-cp313-cp313t-win_arm64.whl", hash = "sha256:bf0a7e10b077bf5fb9380ad3ae8ce20ef919a6ad93b4552896419ac7e1d8e042" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9a/9a/e35b4a917281c0b8419d4207f4334c8e8c5dbf4f3f5f9ada73958d937dcc/frozenlist-1.8.0-py3-none-any.whl", hash = "sha256:0c18a16eab41e82c295618a77502e17b195883241c563b00f0aa5106fc4eaa0d" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "fsspec"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "2026.1.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/d5/7d/5df2650c57d47c57232af5ef4b4fdbff182070421e405e0d62c6cdbfaa87/fsspec-2026.1.0.tar.gz", hash = "sha256:e987cb0496a0d81bba3a9d1cee62922fb395e7d4c3b575e57f547953334fe07b" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/01/c9/97cc5aae1648dcb851958a3ddf73ccd7dbe5650d95203ecb4d7720b4cdbf/fsspec-2026.1.0-py3-none-any.whl", hash = "sha256:cb76aa913c2285a3b49bdd5fc55b1d7c708d7208126b60f2eb8194fe1b4cbdcc" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "future"
version = "1.0.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a7/b2/4140c69c6a66432916b26158687e821ba631a4c9273c474343badf84d3ba/future-1.0.0.tar.gz", hash = "sha256:bd2968309307861edae1458a4f8a4f3598c03be43b97521076aebf5d94c07b05" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/da/71/ae30dadffc90b9006d77af76b393cb9dfbfc9629f339fc1574a1c52e6806/future-1.0.0-py3-none-any.whl", hash = "sha256:929292d34f5872e70396626ef385ec22355a1fae8ad29e1a734c3e43f9fbc216" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "gensim"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "4.4.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "numpy" },
Bump infinity to v0.6.11. Requires python>=3.11 (#11814) ### What problem does this PR solve? Bump infinity to v0.6.11. Requires python>=3.11 ### Type of change - [x] Refactoring
2025-12-09 16:23:37 +08:00
{ name = "scipy" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "smart-open" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/1a/80/fe9d2e1ace968041814dbcfce4e8499a643a36c41267fa4b6c4f54cce420/gensim-4.4.0.tar.gz", hash = "sha256:a3f5b626da5518e79a479140361c663089fe7998df8ba52d56e1ded71ac5bdf5" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/80/6c/4e522973e07ca491d33cc7829996b9e8c8663a16b3f87f580cbdc2732d97/gensim-4.4.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:b8961b7a2bb5190b46bc6cd26c29d5bfea22f99123ed5f506ebd0aaf65996758" },
{ url = "https://mirrors.aliyun.com/pypi/packages/cc/6a/593107ee98331128ed20e5d074865587558a0766659be787a40550ab66df/gensim-4.4.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:59d0d29099a76dd97d4563e002f3488a43e51f99d46387025da38007ebfeeff9" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d9/ef/1675e1a3a04f7d0293a21082f57f4a6a8bf0a9e387da58b71db648b663de/gensim-4.4.0-cp313-cp313-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:3bec3e6a1ecaa6439b21a3e42ceb0ca67ffabc114b646f89b1aab5fe69a39ffc" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b3/b9/ee43ef9c391857232603a9ee281e9c5953f7922d70c98c2296a037d1c0b7/gensim-4.4.0-cp313-cp313-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:9033b18920b7774e68eafacdbd87252ffa29382ec465ddb88bd036e00fc86365" },
{ url = "https://mirrors.aliyun.com/pypi/packages/82/f3/4f8f4d478ce69af812c6002b513c5ad3242976923d172dbe5814903be22f/gensim-4.4.0-cp313-cp313-win_amd64.whl", hash = "sha256:6ecb7aed37fb92d24e15a6adbabe693074003263db0fd9ce97c9f4234a9edc1b" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "google"
version = "3.0.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "beautifulsoup4" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/89/97/b49c69893cddea912c7a660a4b6102c6b02cd268f8c7162dd70b7c16f753/google-3.0.0.tar.gz", hash = "sha256:143530122ee5130509ad5e989f0512f7cb218b2d4eddbafbad40fd10e8d8ccbe" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ac/35/17c9141c4ae21e9a29a43acdfd848e3e468a810517f862cad07977bf8fe9/google-3.0.0-py2.py3-none-any.whl", hash = "sha256:889cf695f84e4ae2c55fbc0cfdaf4c1e729417fa52ab1db0485202ba173e4935" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "google-api-core"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "2.25.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
{ name = "google-auth" },
{ name = "googleapis-common-protos" },
{ name = "proto-plus" },
{ name = "protobuf" },
{ name = "requests" },
Bump infinity to v0.6.11. Requires python>=3.11 (#11814) ### What problem does this PR solve? Bump infinity to v0.6.11. Requires python>=3.11 ### Type of change - [x] Refactoring
2025-12-09 16:23:37 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/09/cd/63f1557235c2440fe0577acdbc32577c5c002684c58c7f4d770a92366a24/google_api_core-2.25.2.tar.gz", hash = "sha256:1c63aa6af0d0d5e37966f157a77f9396d820fba59f9e43e9415bc3dc5baff300" }
Bump infinity to v0.6.11. Requires python>=3.11 (#11814) ### What problem does this PR solve? Bump infinity to v0.6.11. Requires python>=3.11 ### Type of change - [x] Refactoring
2025-12-09 16:23:37 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/c8/d8/894716a5423933f5c8d2d5f04b16f052a515f78e815dab0c2c6f1fd105dc/google_api_core-2.25.2-py3-none-any.whl", hash = "sha256:e9a8f62d363dc8424a8497f4c2a47d6bcda6c16514c935629c257ab5d10210e7" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
build(deps): bump crawl4ai from 0.8.9 to 0.9.0 (#16470)
2026-06-30 09:34:48 +08:00
[package.optional-dependencies]
grpc = [
{ name = "grpcio" },
{ name = "grpcio-status" },
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "google-api-python-client"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "2.197.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
{ name = "google-api-core" },
{ name = "google-auth" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "google-auth-httplib2" },
{ name = "httplib2" },
{ name = "uritemplate" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/22/09/081d66357118bd260f8f182cb1b2dd5bd32ca88e3714d7c93896cab946fc/google_api_python_client-2.197.0.tar.gz", hash = "sha256:32e03977eda4a66eafc6ae58dc9ec46426b6025636d5ef019c5703013eddd4e5" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/a0/e5/e9cc221fd75230974d4ef45eb72d2261feca3c110d5554215d516bfe6534/google_api_python_client-2.197.0-py3-none-any.whl", hash = "sha256:0f8b89aa75768161dd4f5092d6bcb386c13236b32e0d9a938c02f71342094d14" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
[[package]]
name = "google-auth"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "2.41.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
dependencies = [
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
{ name = "cachetools" },
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
{ name = "pyasn1-modules" },
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
{ name = "rsa" },
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a8/af/5129ce5b2f9688d2fa49b463e544972a7c82b0fdb50980dafee92e121d9f/google_auth-2.41.1.tar.gz", hash = "sha256:b76b7b1f9e61f0cb7e88870d14f6a94aeef248959ef6992670efee37709cbfd2" }
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/be/a4/7319a2a8add4cc352be9e3efeff5e2aacee917c85ca2fa1647e29089983c/google_auth-2.41.1-py2.py3-none-any.whl", hash = "sha256:754843be95575b9a19c604a848a41be03f7f2afd8c019f716dc1f51ee41c639d" },
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
]
[package.optional-dependencies]
build(deps): bump crawl4ai from 0.8.9 to 0.9.0 (#16470)
2026-06-30 09:34:48 +08:00
pyopenssl = [
{ name = "cryptography" },
{ name = "pyopenssl" },
]
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
requests = [
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
{ name = "requests" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "google-auth-httplib2"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "0.3.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
{ name = "google-auth" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "httplib2" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/d5/ad/c1f2b1175096a8d04cf202ad5ea6065f108d26be6fc7215876bde4a7981d/google_auth_httplib2-0.3.0.tar.gz", hash = "sha256:177898a0175252480d5ed916aeea183c2df87c1f9c26705d74ae6b951c268b0b" }
Feat: add initial Google Drive connector support (#11147) ### What problem does this PR solve? This feature is primarily ported from the [Onyx](https://github.com/onyx-dot-app/onyx) project with necessary modifications. Thanks for such a brilliant project. Minor: consistently use `google_drive` rather than `google_driver`. <img width="566" height="731" alt="image" src="https://github.com/user-attachments/assets/6f64e70e-881e-42c7-b45f-809d3e0024a4" /> <img width="904" height="830" alt="image" src="https://github.com/user-attachments/assets/dfa7d1ef-819a-4a82-8c52-0999f48ed4a6" /> <img width="911" height="869" alt="image" src="https://github.com/user-attachments/assets/39e792fb-9fbe-4f3d-9b3c-b2265186bc22" /> <img width="947" height="323" alt="image" src="https://github.com/user-attachments/assets/27d70e96-d9c0-42d9-8c89-276919b6d61d" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-10 19:15:02 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/99/d5/3c97526c8796d3caf5f4b3bed2b05e8a7102326f00a334e7a438237f3b22/google_auth_httplib2-0.3.0-py3-none-any.whl", hash = "sha256:426167e5df066e3f5a0fc7ea18768c08e7296046594ce4c8c409c2457dd1f776" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
[[package]]
name = "google-auth-oauthlib"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.2.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
dependencies = [
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
{ name = "google-auth" },
{ name = "requests-oauthlib" },
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/86/a6/c6336a6ceb682709a4aa39e2e6b5754a458075ca92359512b6cbfcb25ae3/google_auth_oauthlib-1.2.3.tar.gz", hash = "sha256:eb09e450d3cc789ecbc2b3529cb94a713673fd5f7a22c718ad91cf75aedc2ea4" }
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/38/07/a54c100da461ffc5968457823fcc665a48fb4b875c68bcfecbfe24a10dbe/google_auth_oauthlib-1.2.3-py3-none-any.whl", hash = "sha256:7c0940e037677f25e71999607493640d071212e7f3c15aa0febea4c47a5a0680" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
build(deps): bump crawl4ai from 0.8.9 to 0.9.0 (#16470)
2026-06-30 09:34:48 +08:00
[[package]]
name = "google-cloud-bigquery"
version = "3.42.1"
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
dependencies = [
{ name = "google-api-core", extra = ["grpc"] },
{ name = "google-auth", extra = ["pyopenssl"] },
{ name = "google-cloud-core" },
{ name = "google-resumable-media" },
{ name = "packaging" },
{ name = "python-dateutil" },
{ name = "requests" },
]
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/15/a2/be1cffbb1a9894ecf394ab0096b113ae9e4f73686595d906a9f3082b07c2/google_cloud_bigquery-3.42.1.tar.gz", hash = "sha256:3c6878f424fc2b21f0bb414ca7262abd008465575a7b7c44c552105278a18914" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/7a/1e/8b098aeb69ef87de5195d076ce6d82002c3ebf1920373d386ca7b6e63cfa/google_cloud_bigquery-3.42.1-py3-none-any.whl", hash = "sha256:159143c1b7248858f35549c09efea5d031c2883145623419ea7c505a50c0899a" },
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "google-cloud-core"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "2.5.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
{ name = "google-api-core" },
{ name = "google-auth" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a6/03/ef0bc99d0e0faf4fdbe67ac445e18cdaa74824fd93cd069e7bb6548cb52d/google_cloud_core-2.5.0.tar.gz", hash = "sha256:7c1b7ef5c92311717bd05301aa1a91ffbc565673d3b0b4163a52d8413a186963" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/89/20/bfa472e327c8edee00f04beecc80baeddd2ab33ee0e86fd7654da49d45e9/google_cloud_core-2.5.0-py3-none-any.whl", hash = "sha256:67d977b41ae6c7211ee830c7912e41003ea8194bff15ae7d72fd6f51e57acabc" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "google-cloud-storage"
version = "2.19.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
{ name = "google-api-core" },
{ name = "google-auth" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "google-cloud-core" },
{ name = "google-crc32c" },
{ name = "google-resumable-media" },
{ name = "requests" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/36/76/4d965702e96bb67976e755bed9828fa50306dca003dbee08b67f41dd265e/google_cloud_storage-2.19.0.tar.gz", hash = "sha256:cd05e9e7191ba6cb68934d8eb76054d9be4562aa89dbc4236feee4d7d51342b2" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/d5/94/6db383d8ee1adf45dc6c73477152b82731fa4c4a46d9c1932cc8757e0fd4/google_cloud_storage-2.19.0-py2.py3-none-any.whl", hash = "sha256:aeb971b5c29cf8ab98445082cbfe7b161a1f48ed275822f59ed3f1524ea54fba" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "google-crc32c"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "1.8.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/03/41/4b9c02f99e4c5fb477122cd5437403b552873f014616ac1d19ac8221a58d/google_crc32c-1.8.0.tar.gz", hash = "sha256:a428e25fb7691024de47fecfbff7ff957214da51eddded0da0ae0e0f03a2cf79" }
Fix: switch Python package mirror from Tsinghua to Aliyun (#13617) ### What problem does this PR solve? Replace pypi.tuna.tsinghua.edu.cn with mirrors.aliyun.com to resolve issues with missing packages on the Tsinghua mirror. ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2026-03-16 12:12:25 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/d1/db/000f15b41724589b0e7bc24bc7a8967898d8d3bc8caf64c513d91ef1f6c0/google_crc32c-1.8.0-cp313-cp313-macosx_12_0_arm64.whl", hash = "sha256:3ebb04528e83b2634857f43f9bb8ef5b2bbe7f10f140daeb01b58f972d04736b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d7/0d/8ebed0c39c53a7e838e2a486da8abb0e52de135f1b376ae2f0b160eb4c1a/google_crc32c-1.8.0-cp313-cp313-macosx_12_0_x86_64.whl", hash = "sha256:450dc98429d3e33ed2926fc99ee81001928d63460f8538f21a5d6060912a8e27" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ce/42/b468aec74a0354b34c8cbf748db20d6e350a68a2b0912e128cabee49806c/google_crc32c-1.8.0-cp313-cp313-manylinux1_x86_64.manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:3b9776774b24ba76831609ffbabce8cdf6fa2bd5e9df37b594221c7e333a81fa" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1c/e8/b33784d6fc77fb5062a8a7854e43e1e618b87d5ddf610a88025e4de6226e/google_crc32c-1.8.0-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:89c17d53d75562edfff86679244830599ee0a48efc216200691de8b02ab6b2b8" },
{ url = "https://mirrors.aliyun.com/pypi/packages/92/b1/d3cbd4d988afb3d8e4db94ca953df429ed6db7282ed0e700d25e6c7bfc8d/google_crc32c-1.8.0-cp313-cp313-win_amd64.whl", hash = "sha256:57a50a9035b75643996fbf224d6661e386c7162d1dfdab9bc4ca790947d1007f" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Fix problem with Google Cloud models with reasoning (like gemini) - Additional fix to issue #10474 (#10502) ### What problem does this PR solve? Issue #10474 - Update to PR #10477 ### Type of change - [X] Bug Fix (non-breaking change which fixes an issue)
2025-10-15 08:54:20 +02:00
[[package]]
name = "google-genai"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.55.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Fix problem with Google Cloud models with reasoning (like gemini) - Additional fix to issue #10474 (#10502) ### What problem does this PR solve? Issue #10474 - Update to PR #10477 ### Type of change - [X] Bug Fix (non-breaking change which fixes an issue)
2025-10-15 08:54:20 +02:00
dependencies = [
{ name = "anyio" },
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
{ name = "distro" },
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
{ name = "google-auth", extra = ["requests"] },
Fix problem with Google Cloud models with reasoning (like gemini) - Additional fix to issue #10474 (#10502) ### What problem does this PR solve? Issue #10474 - Update to PR #10477 ### Type of change - [X] Bug Fix (non-breaking change which fixes an issue)
2025-10-15 08:54:20 +02:00
{ name = "httpx" },
{ name = "pydantic" },
{ name = "requests" },
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
{ name = "sniffio" },
Fix problem with Google Cloud models with reasoning (like gemini) - Additional fix to issue #10474 (#10502) ### What problem does this PR solve? Issue #10474 - Update to PR #10477 ### Type of change - [X] Bug Fix (non-breaking change which fixes an issue)
2025-10-15 08:54:20 +02:00
{ name = "tenacity" },
{ name = "typing-extensions" },
{ name = "websockets" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/1d/7c/19b59750592702305ae211905985ec8ab56f34270af4a159fba5f0214846/google_genai-1.55.0.tar.gz", hash = "sha256:ae9f1318fedb05c7c1b671a4148724751201e8908a87568364a309804064d986" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/3e/86/a5a8e32b2d40b30b5fb20e7b8113fafd1e38befa4d1801abd5ce6991065a/google_genai-1.55.0-py3-none-any.whl", hash = "sha256:98c422762b5ff6e16b8d9a1e4938e8e0ad910392a5422e47f5301498d7f373a1" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "google-resumable-media"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "2.8.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "google-crc32c" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/64/d7/520b62a35b23038ff005e334dba3ffc75fcf583bee26723f1fd8fd4b6919/google_resumable_media-2.8.0.tar.gz", hash = "sha256:f1157ed8b46994d60a1bc432544db62352043113684d4e030ee02e77ebe9a1ae" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/1f/0b/93afde9cfe012260e9fe1522f35c9b72d6ee222f316586b1f23ecf44d518/google_resumable_media-2.8.0-py3-none-any.whl", hash = "sha256:dd14a116af303845a8d932ddae161a26e86cc229645bc98b39f026f9b1717582" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "google-search-results"
version = "2.4.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "requests" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/77/30/b3a6f6a2e00f8153549c2fa345c58ae1ce8e5f3153c2fe0484d444c3abcb/google_search_results-2.4.2.tar.gz", hash = "sha256:603a30ecae2af8e600b22635757a6df275dad4b934f975e67878ccd640b78245" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "googleapis-common-protos"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.72.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "protobuf" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/e5/7b/adfd75544c415c487b33061fe7ae526165241c1ea133f9a9125a56b39fd8/googleapis_common_protos-1.72.0.tar.gz", hash = "sha256:e55a601c1b32b52d7a3e65f43563e2aa61bcd737998ee672ac9b951cd49319f5" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/c4/ab/09169d5a4612a5f92490806649ac8d41e3ec9129c636754575b3553f4ea4/googleapis_common_protos-1.72.0-py3-none-any.whl", hash = "sha256:4299c5a82d5ae1a9702ada957347726b167f9f8d1fc352477702a1e851ff4038" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "gprofiler-official"
version = "1.0.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "requests" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ec/c1/d9252620d09a064247d1623ebc4732d624921a2ed80a677f8b9ce61810dd/gprofiler-official-1.0.0.tar.gz", hash = "sha256:5015b47f10fbdcb59c57e342e815c9c07afbe57cd3984154f75b845ddef2445d" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/df/1b/5a87c1a1da8f601c00a0ce4dedb5aab8a5cad6a0f4a5062c4da22a045072/gprofiler_official-1.0.0-py3-none-any.whl", hash = "sha256:c582baf728e5a6cddac964e4085ca385e082c4ef0279e3af1a16a9af07ab5395" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "graspologic"
Bump beartype to 0.22.6 (#11463) ### What problem does this PR solve? Bump beartype to 0.22.6 ### Type of change - [x] Refactoring
2025-11-22 11:56:43 +08:00
version = "0.1.dev847+g38e680cab"
Update graspologic to gitee (#13362) ### What problem does this PR solve? Accelerate python module downloading ### Type of change - [x] Refactoring Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2026-03-04 17:48:47 +08:00
source = { git = "https://gitee.com/infiniflow/graspologic.git?rev=38e680cab72bc9fb68a7992c3bcc2d53b24e42fd#38e680cab72bc9fb68a7992c3bcc2d53b24e42fd" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "anytree" },
{ name = "beartype" },
Bump beartype to 0.22.6 (#11463) ### What problem does this PR solve? Bump beartype to 0.22.6 ### Type of change - [x] Refactoring
2025-11-22 11:56:43 +08:00
{ name = "future" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "gensim" },
{ name = "graspologic-native" },
{ name = "hyppo" },
{ name = "joblib" },
{ name = "matplotlib" },
Bump infinity to v0.6.11. Requires python>=3.11 (#11814) ### What problem does this PR solve? Bump infinity to v0.6.11. Requires python>=3.11 ### Type of change - [x] Refactoring
2025-12-09 16:23:37 +08:00
{ name = "networkx" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "numpy" },
{ name = "pot" },
{ name = "scikit-learn" },
Bump infinity to v0.6.11. Requires python>=3.11 (#11814) ### What problem does this PR solve? Bump infinity to v0.6.11. Requires python>=3.11 ### Type of change - [x] Refactoring
2025-12-09 16:23:37 +08:00
{ name = "scipy" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "seaborn" },
{ name = "statsmodels" },
{ name = "typing-extensions" },
{ name = "umap-learn" },
]
[[package]]
name = "graspologic-native"
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
version = "1.2.5"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/25/2d/62b30d89533643ccf4778a18eb023f291b8877b5d85de3342f07b2d363a7/graspologic_native-1.2.5.tar.gz", hash = "sha256:27ea7e01fa44466c0b4cdd678d4561e5d3dc0cb400015683b7ae1386031257a0" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ae/86/10748f4c474b0c8f6060dd379bb0c4da5d42779244bb13a58656ffb44a03/graspologic_native-1.2.5-cp38-abi3-macosx_10_12_x86_64.macosx_11_0_arm64.macosx_10_12_universal2.whl", hash = "sha256:bf05f2e162ae2a2a8d6e8cfccbe3586d1faa0b808159ff950478348df557c61e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/42/cc/b75ea35755340bedda29727e5388390c639ea533f55b9249f5ac3003f656/graspologic_native-1.2.5-cp38-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4a7fff06ed49c3875cf351bb09a92ae7cbc169ce92dcc4c3439e28e801f822ae" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8e/55/15e6e4f18bf249b529ac4cd1522b03f5c9ef9284a2f7bfaa1fd1f96464fe/graspologic_native-1.2.5-cp38-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:53e7e993e7d70fe0d860773fc62812fbb8cb4ef2d11d8661a1f06f8772593915" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3b/51/21097af79f3d68626539ab829bdbf6cc42933f020e161972927d916e394c/graspologic_native-1.2.5-cp38-abi3-win_amd64.whl", hash = "sha256:c3ef2172d774083d7e2c8e77daccd218571ddeebeb2c1703cebb1a2cc4c56e07" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "greenlet"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "3.3.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/c7/e5/40dbda2736893e3e53d25838e0f19a2b417dfc122b9989c91918db30b5d3/greenlet-3.3.0.tar.gz", hash = "sha256:a82bb225a4e9e4d653dd2fb7b8b2d36e4fb25bc0165422a11e48b88e9e6f78fb" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/02/2f/28592176381b9ab2cafa12829ba7b472d177f3acc35d8fbcf3673d966fff/greenlet-3.3.0-cp313-cp313-macosx_11_0_universal2.whl", hash = "sha256:a1e41a81c7e2825822f4e068c48cb2196002362619e2d70b148f20a831c00739" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2c/80/fbe937bf81e9fca98c981fe499e59a3f45df2a04da0baa5c2be0dca0d329/greenlet-3.3.0-cp313-cp313-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:9f515a47d02da4d30caaa85b69474cec77b7929b2e936ff7fb853d42f4bf8808" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c2/ff/7c985128f0514271b8268476af89aee6866df5eec04ac17dcfbc676213df/greenlet-3.3.0-cp313-cp313-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:7d2d9fd66bfadf230b385fdc90426fcd6eb64db54b40c495b72ac0feb5766c54" },
{ url = "https://mirrors.aliyun.com/pypi/packages/79/07/c47a82d881319ec18a4510bb30463ed6891f2ad2c1901ed5ec23d3de351f/greenlet-3.3.0-cp313-cp313-manylinux_2_24_s390x.manylinux_2_28_s390x.whl", hash = "sha256:30a6e28487a790417d036088b3bcb3f3ac7d8babaa7d0139edbaddebf3af9492" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fd/8e/424b8c6e78bd9837d14ff7df01a9829fc883ba2ab4ea787d4f848435f23f/greenlet-3.3.0-cp313-cp313-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:087ea5e004437321508a8d6f20efc4cfec5e3c30118e1417ea96ed1d93950527" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b5/ba/56699ff9b7c76ca12f1cdc27a886d0f81f2189c3455ff9f65246780f713d/greenlet-3.3.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:ab97cf74045343f6c60a39913fa59710e4bd26a536ce7ab2397adf8b27e67c39" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1e/37/f31136132967982d698c71a281a8901daf1a8fbab935dce7c0cf15f942cc/greenlet-3.3.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:5375d2e23184629112ca1ea89a53389dddbffcf417dad40125713d88eb5f96e8" },
{ url = "https://mirrors.aliyun.com/pypi/packages/7e/71/ba21c3fb8c5dce83b8c01f458a42e99ffdb1963aeec08fff5a18588d8fd7/greenlet-3.3.0-cp313-cp313-win_amd64.whl", hash = "sha256:9ee1942ea19550094033c35d25d20726e4f1c40d59545815e1128ac58d416d38" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "groq"
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
version = "0.37.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "anyio" },
{ name = "distro" },
{ name = "httpx" },
{ name = "pydantic" },
{ name = "sniffio" },
{ name = "typing-extensions" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/e9/78/18948a9056e1509c87e10ab8316a90ecce87035fbd53342dffdf97f4de00/groq-0.37.1.tar.gz", hash = "sha256:7353d6dfb60834fd7aacbb86af106e2dc2aeaff6d0edd65fb2fd0f16bd39314c" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/5f/d6/645a081750e43f858b7d09dce5d8e1e76cf11e7e4bdba81252e04f78963d/groq-0.37.1-py3-none-any.whl", hash = "sha256:b49f8c8898c55eaec9f71f1342f3fcacc9560d67a08ce5f35fbfb84e8dacd3da" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "grpcio"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.76.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
dependencies = [
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
{ name = "typing-extensions" },
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/b6/e0/318c1ce3ae5a17894d5791e87aea147587c9e702f24122cc7a5c8bbaeeb1/grpcio-1.76.0.tar.gz", hash = "sha256:7be78388d6da1a25c0d5ec506523db58b18be22d9c37d8d3a32c08be4987bd73" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/fc/ed/71467ab770effc9e8cef5f2e7388beb2be26ed642d567697bb103a790c72/grpcio-1.76.0-cp313-cp313-linux_armv7l.whl", hash = "sha256:26ef06c73eb53267c2b319f43e6634c7556ea37672029241a056629af27c10e2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2c/85/c6ed56f9817fab03fa8a111ca91469941fb514e3e3ce6d793cb8f1e1347b/grpcio-1.76.0-cp313-cp313-macosx_11_0_universal2.whl", hash = "sha256:45e0111e73f43f735d70786557dc38141185072d7ff8dc1829d6a77ac1471468" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ac/31/2b8a235ab40c39cbc141ef647f8a6eb7b0028f023015a4842933bc0d6831/grpcio-1.76.0-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:83d57312a58dcfe2a3a0f9d1389b299438909a02db60e2f2ea2ae2d8034909d3" },
{ url = "https://mirrors.aliyun.com/pypi/packages/bd/64/9784eab483358e08847498ee56faf8ff6ea8e0a4592568d9f68edc97e9e9/grpcio-1.76.0-cp313-cp313-manylinux2014_i686.manylinux_2_17_i686.whl", hash = "sha256:3e2a27c89eb9ac3d81ec8835e12414d73536c6e620355d65102503064a4ed6eb" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2b/94/8c12319a6369434e7a184b987e8e9f3b49a114c489b8315f029e24de4837/grpcio-1.76.0-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:61f69297cba3950a524f61c7c8ee12e55c486cb5f7db47ff9dcee33da6f0d3ae" },
{ url = "https://mirrors.aliyun.com/pypi/packages/15/0f/f12c32b03f731f4a6242f771f63039df182c8b8e2cf8075b245b409259d4/grpcio-1.76.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:6a15c17af8839b6801d554263c546c69c4d7718ad4321e3166175b37eaacca77" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ff/2d/3ec9ce0c2b1d92dd59d1c3264aaec9f0f7c817d6e8ac683b97198a36ed5a/grpcio-1.76.0-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:25a18e9810fbc7e7f03ec2516addc116a957f8cbb8cbc95ccc80faa072743d03" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1a/74/fd3317be5672f4856bcdd1a9e7b5e17554692d3db9a3b273879dc02d657d/grpcio-1.76.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:931091142fd8cc14edccc0845a79248bc155425eee9a98b2db2ea4f00a235a42" },
{ url = "https://mirrors.aliyun.com/pypi/packages/45/bb/ca038cf420f405971f19821c8c15bcbc875505f6ffadafe9ffd77871dc4c/grpcio-1.76.0-cp313-cp313-win32.whl", hash = "sha256:5e8571632780e08526f118f74170ad8d50fb0a48c23a746bef2a6ebade3abd6f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/41/80/84087dc56437ced7cdd4b13d7875e7439a52a261e3ab4e06488ba6173b0a/grpcio-1.76.0-cp313-cp313-win_amd64.whl", hash = "sha256:f9f7bd5faab55f47231ad8dba7787866b69f5e93bc306e3915606779bbfb4ba8" },
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
]
[[package]]
name = "grpcio-status"
version = "1.67.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
dependencies = [
Remove unused py module dependencies (#11964) ### What problem does this PR solve? ``` pip list --not-required Package Version ---------------------------- --------------------- aiosmtplib 5.0.0 akshare 1.17.94 anthropic 0.34.1 arxiv 2.1.3 Aspose.Slides 24.7.0 atlassian-python-api 4.0.7 azure-identity 1.17.1 azure-storage-file-datalake 12.16.0 bio 1.7.1 boxsdk 10.2.0 captcha 0.7.1 cn2an 0.5.22 cohere 5.6.2 Crawl4AI 0.4.247 dashscope 1.20.11 deepl 1.18.0 demjson3 3.0.6 discord.py 2.3.2 dropbox 12.0.2 duckduckgo_search 7.5.5 editdistance 0.8.1 elasticsearch-dsl 8.12.0 exceptiongroup 1.3.1 extract-msg 0.55.0 ffmpeg-python 0.2.0 flasgger 0.9.7.1 Flask-Cors 5.0.0 Flask-Login 0.6.3 Flask-Mail 0.10.0 Flask-Session 0.8.0 google-auth-oauthlib 1.2.3 google-genai 1.55.0 google-generativeai 0.8.5 google_search_results 2.4.2 graspologic 0.1.dev847+g38e680cab groq 0.9.0 grpcio-status 1.67.1 html_text 0.6.2 imageio-ffmpeg 0.6.0 infinity_emb 0.0.66 infinity-sdk 0.6.11 jira 3.10.5 json_repair 0.35.0 langfuse 3.10.5 mammoth 1.11.0 Markdown 3.6 markdown_to_json 2.1.1 markdownify 1.2.2 mcp 1.19.0 mini-racer 0.12.4 minio 7.2.4 mistralai 0.4.2 moodlepy 0.24.1 mypy-boto3-s3 1.40.26 Office365-REST-Python-Client 2.6.2 ollama 0.6.1 onnxruntime-gpu 1.23.2 opencv-python 4.10.0.84 opencv-python-headless 4.10.0.84 opendal 0.45.20 opensearch-py 2.7.1 ormsgpack 1.5.0 pdfplumber 0.10.4 pip 25.3 pluginlib 0.9.4 psycopg2-binary 2.9.11 pyclipper 1.4.0 pycryptodomex 3.20.0 pyobvector 0.2.18 pyodbc 5.3.0 pypandoc 1.16.2 pypdf 6.4.0 PyPDF2 3.0.1 python-calamine 0.6.1 python-docx 1.2.0 python-pptx 1.0.2 pywencai 0.13.1 qianfan 0.4.6 quart-auth 0.11.0 quart-cors 0.8.0 ranx 0.3.20 readability-lxml 0.8.4.1 replicate 0.31.0 reportlab 4.4.6 roman-numbers 1.0.2 ruamel.base 1.0.0 ruamel.yaml 0.18.16 scholarly 1.7.11 selenium-wire 5.1.0 slack_sdk 3.37.0 socksio 1.0.0 sqlglotrs 0.9.0 StrEnum 0.4.15 tavily-python 0.5.1 tencentcloud-sdk-python 3.0.1478 tika 2.6.0 valkey 6.0.2 vertexai 1.70.0 volcengine 1.0.194 voyageai 0.2.3 webdav4 0.10.0 webdriver-manager 4.0.1 wikipedia 1.4.0 word2number 1.1 xgboost 1.6.0 xpinyin 0.7.6 yfinance 0.2.65 zhipuai 2.0.1 ``` ### Type of change - [x] Refactoring Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-16 12:40:03 +08:00
{ name = "googleapis-common-protos" },
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
{ name = "grpcio" },
Remove unused py module dependencies (#11964) ### What problem does this PR solve? ``` pip list --not-required Package Version ---------------------------- --------------------- aiosmtplib 5.0.0 akshare 1.17.94 anthropic 0.34.1 arxiv 2.1.3 Aspose.Slides 24.7.0 atlassian-python-api 4.0.7 azure-identity 1.17.1 azure-storage-file-datalake 12.16.0 bio 1.7.1 boxsdk 10.2.0 captcha 0.7.1 cn2an 0.5.22 cohere 5.6.2 Crawl4AI 0.4.247 dashscope 1.20.11 deepl 1.18.0 demjson3 3.0.6 discord.py 2.3.2 dropbox 12.0.2 duckduckgo_search 7.5.5 editdistance 0.8.1 elasticsearch-dsl 8.12.0 exceptiongroup 1.3.1 extract-msg 0.55.0 ffmpeg-python 0.2.0 flasgger 0.9.7.1 Flask-Cors 5.0.0 Flask-Login 0.6.3 Flask-Mail 0.10.0 Flask-Session 0.8.0 google-auth-oauthlib 1.2.3 google-genai 1.55.0 google-generativeai 0.8.5 google_search_results 2.4.2 graspologic 0.1.dev847+g38e680cab groq 0.9.0 grpcio-status 1.67.1 html_text 0.6.2 imageio-ffmpeg 0.6.0 infinity_emb 0.0.66 infinity-sdk 0.6.11 jira 3.10.5 json_repair 0.35.0 langfuse 3.10.5 mammoth 1.11.0 Markdown 3.6 markdown_to_json 2.1.1 markdownify 1.2.2 mcp 1.19.0 mini-racer 0.12.4 minio 7.2.4 mistralai 0.4.2 moodlepy 0.24.1 mypy-boto3-s3 1.40.26 Office365-REST-Python-Client 2.6.2 ollama 0.6.1 onnxruntime-gpu 1.23.2 opencv-python 4.10.0.84 opencv-python-headless 4.10.0.84 opendal 0.45.20 opensearch-py 2.7.1 ormsgpack 1.5.0 pdfplumber 0.10.4 pip 25.3 pluginlib 0.9.4 psycopg2-binary 2.9.11 pyclipper 1.4.0 pycryptodomex 3.20.0 pyobvector 0.2.18 pyodbc 5.3.0 pypandoc 1.16.2 pypdf 6.4.0 PyPDF2 3.0.1 python-calamine 0.6.1 python-docx 1.2.0 python-pptx 1.0.2 pywencai 0.13.1 qianfan 0.4.6 quart-auth 0.11.0 quart-cors 0.8.0 ranx 0.3.20 readability-lxml 0.8.4.1 replicate 0.31.0 reportlab 4.4.6 roman-numbers 1.0.2 ruamel.base 1.0.0 ruamel.yaml 0.18.16 scholarly 1.7.11 selenium-wire 5.1.0 slack_sdk 3.37.0 socksio 1.0.0 sqlglotrs 0.9.0 StrEnum 0.4.15 tavily-python 0.5.1 tencentcloud-sdk-python 3.0.1478 tika 2.6.0 valkey 6.0.2 vertexai 1.70.0 volcengine 1.0.194 voyageai 0.2.3 webdav4 0.10.0 webdriver-manager 4.0.1 wikipedia 1.4.0 word2number 1.1 xgboost 1.6.0 xpinyin 0.7.6 yfinance 0.2.65 zhipuai 2.0.1 ``` ### Type of change - [x] Refactoring Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-16 12:40:03 +08:00
{ name = "protobuf" },
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/be/c7/fe0e79a80ac6346e0c6c0a24e9e3cbc3ae1c2a009acffb59eab484a6f69b/grpcio_status-1.67.1.tar.gz", hash = "sha256:2bf38395e028ceeecfd8866b081f61628114b384da7d51ae064ddc8d766a5d11" }
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/05/18/56999a1da3577d8ccc8698a575d6638e15fe25650cc88b2ce0a087f180b9/grpcio_status-1.67.1-py3-none-any.whl", hash = "sha256:16e6c085950bdacac97c779e6a502ea671232385e6e37f258884d6883392c2bd" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "h11"
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
version = "0.16.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/01/ee/02a2c011bdab74c6fb3c75474d40b3052059d95df7e73351460c8588d963/h11-0.16.0.tar.gz", hash = "sha256:4e35b956cf45792e4caa5885e69fba00bdbc6ffafbfa020300e549b208ee5ff1" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/04/4b/29cac41a4d98d144bf5f6d33995617b185d14b22401f75ca86f384e87ff1/h11-0.16.0-py3-none-any.whl", hash = "sha256:63cf8bbe7522de3bf65932fda1d9c2772064ffb3dae62d55932da54b31cb6c86" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "h2"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "4.3.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "hpack" },
{ name = "hyperframe" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/1d/17/afa56379f94ad0fe8defd37d6eb3f89a25404ffc71d4d848893d270325fc/h2-4.3.0.tar.gz", hash = "sha256:6c59efe4323fa18b47a632221a1888bd7fde6249819beda254aeca909f221bf1" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/69/b2/119f6e6dcbd96f9069ce9a2665e0146588dc9f88f29549711853645e736a/h2-4.3.0-py3-none-any.whl", hash = "sha256:c438f029a25f7945c69e0ccf0fb951dc3f73a5f6412981daee861431b70e2bdd" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "hanziconv"
version = "0.3.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/63/71/b89cb63077fd807fe31cf7c016a06e7e579a289d8a37aa24a30282d02dd2/hanziconv-0.3.2.tar.gz", hash = "sha256:208866da6ae305bca19eb98702b65c93bb3a803b496e4287ca740d68892fc4c4" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "hf-transfer"
version = "0.1.9"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/1a/eb/8fc64f40388c29ce8ce3b2b180a089d4d6b25b1d0d232d016704cb852104/hf_transfer-0.1.9.tar.gz", hash = "sha256:035572865dab29d17e783fbf1e84cf1cb24f3fcf8f1b17db1cfc7fdf139f02bf" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/a4/78/0dce00208f585fae675f40033ef9a30dedfa83665d5ac79f16beb4a0a6c2/hf_transfer-0.1.9-cp313-cp313t-macosx_10_12_x86_64.whl", hash = "sha256:6e94e8822da79573c9b6ae4d6b2f847c59a7a06c5327d7db20751b68538dc4f6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ea/2e/3d60b1a9e9f29a2152aa66c823bf5e399ae7be3fef310ff0de86779c5d2d/hf_transfer-0.1.9-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:3ebc4ab9023414880c8b1d3c38174d1c9989eb5022d37e814fa91a3060123eb0" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fb/38/130a5ac3747f104033591bcac1c961cb1faadfdc91704f59b09c0b465ff2/hf_transfer-0.1.9-cp313-cp313t-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8674026f21ed369aa2a0a4b46000aca850fc44cd2b54af33a172ce5325b4fc82" },
{ url = "https://mirrors.aliyun.com/pypi/packages/15/a1/f4e27c5ad17aac616ae0849e2aede5aae31db8267a948c6b3eeb9fd96446/hf_transfer-0.1.9-cp313-cp313t-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:3a736dfbb2c84f5a2c975478ad200c0c8bfcb58a25a35db402678fb87ce17fa4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8d/0d/727abdfba39bc3f1132cfa4c970588c2c0bb0d82fe2d645cc10f4e2f8e0b/hf_transfer-0.1.9-cp313-cp313t-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:504b8427fd785dd8546d53b9fafe6e436bd7a3adf76b9dce556507650a7b4567" },
{ url = "https://mirrors.aliyun.com/pypi/packages/50/d0/2b213eb1ea8b1252ccaf1a6c804d0aba03fea38aae4124df6a3acb70511a/hf_transfer-0.1.9-cp313-cp313t-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:2c7fc1b85f4d0f76e452765d7648c9f4bfd0aedb9ced2ae1ebfece2d8cfaf8e2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8c/8a/79dbce9006e0bd6b74516f97451a7b7c64dbbb426df15d901dd438cfeee3/hf_transfer-0.1.9-cp313-cp313t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0d991376f0eac70a60f0cbc95602aa708a6f7c8617f28b4945c1431d67b8e3c8" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a9/f7/9ac239b6ee6fe0bad130325d987a93ea58c4118e50479f0786f1733b37e8/hf_transfer-0.1.9-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:e6ac4eddcd99575ed3735ed911ddf9d1697e2bd13aa3f0ad7e3904dd4863842e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d8/a3/0ed697279f5eeb7a40f279bd783cf50e6d0b91f24120dcf66ef2cf8822b4/hf_transfer-0.1.9-cp313-cp313t-musllinux_1_2_armv7l.whl", hash = "sha256:57fd9880da1ee0f47250f735f791fab788f0aa1ee36afc49f761349869c8b4d9" },
{ url = "https://mirrors.aliyun.com/pypi/packages/dc/eb/47e477bdf1d784f31c7540db6cc8c354b777e51a186897a7abda34517f36/hf_transfer-0.1.9-cp313-cp313t-musllinux_1_2_i686.whl", hash = "sha256:5d561f0520f493c66b016d99ceabe69c23289aa90be38dd802d2aef279f15751" },
{ url = "https://mirrors.aliyun.com/pypi/packages/45/07/6661e43fbee09594a8a5e9bb778107d95fe38dac4c653982afe03d32bd4d/hf_transfer-0.1.9-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:a5b366d34cd449fe9b20ef25941e6eef0460a2f74e7389f02e673e1f88ebd538" },
{ url = "https://mirrors.aliyun.com/pypi/packages/81/f5/461d2e5f307e5048289b1168d5c642ae3bb2504e88dff1a38b92ed990a21/hf_transfer-0.1.9-cp38-abi3-macosx_10_12_x86_64.whl", hash = "sha256:e66acf91df4a8b72f60223059df3003062a5ae111757187ed1a06750a30e911b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/41/ba/8d9fd9f1083525edfcb389c93738c802f3559cb749324090d7109c8bf4c2/hf_transfer-0.1.9-cp38-abi3-macosx_11_0_arm64.whl", hash = "sha256:8669dbcc7a3e2e8d61d42cd24da9c50d57770bd74b445c65123291ca842a7e7a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8e/a2/cd7885bc9959421065a6fae0fe67b6c55becdeda4e69b873e52976f9a9f0/hf_transfer-0.1.9-cp38-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:8fd0167c4407a3bc4cdd0307e65ada2294ec04f1813d8a69a5243e379b22e9d8" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f6/2e/a072cf196edfeda3310c9a5ade0a0fdd785e6154b3ce24fc738c818da2a7/hf_transfer-0.1.9-cp38-abi3-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:ee8b10afedcb75f71091bcc197c526a6ebf5c58bbbadb34fdeee6160f55f619f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c2/84/aec9ef4c0fab93c1ea2b1badff38c78b4b2f86f0555b26d2051dbc920cde/hf_transfer-0.1.9-cp38-abi3-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:5828057e313de59300dd1abb489444bc452efe3f479d3c55b31a8f680936ba42" },
{ url = "https://mirrors.aliyun.com/pypi/packages/29/63/b560d39651a56603d64f1a0212d0472a44cbd965db2fa62b99d99cb981bf/hf_transfer-0.1.9-cp38-abi3-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:fc6bd19e1cc177c66bdef15ef8636ad3bde79d5a4f608c158021153b4573509d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d6/d8/f87ea6f42456254b48915970ed98e993110521e9263472840174d32c880d/hf_transfer-0.1.9-cp38-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:cdca9bfb89e6f8f281890cc61a8aff2d3cecaff7e1a4d275574d96ca70098557" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d6/56/1267c39b65fc8f4e2113b36297320f102718bf5799b544a6cbe22013aa1d/hf_transfer-0.1.9-cp38-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:89a23f58b7b7effbc047b8ca286f131b17728c99a9f972723323003ffd1bb916" },
{ url = "https://mirrors.aliyun.com/pypi/packages/82/1a/9c748befbe3decf7cb415e34f8a0c3789a0a9c55910dea73d581e48c0ce5/hf_transfer-0.1.9-cp38-abi3-musllinux_1_2_armv7l.whl", hash = "sha256:dc7fff1345980d6c0ebb92c811d24afa4b98b3e07ed070c8e38cc91fd80478c5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/72/85/4c03da147b6b4b7cb12e074d3d44eee28604a387ed0eaf7eaaead5069c57/hf_transfer-0.1.9-cp38-abi3-musllinux_1_2_i686.whl", hash = "sha256:1a6bd16c667ebe89a069ca163060127a794fa3a3525292c900b8c8cc47985b0d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e7/6e/e597b04f753f1b09e6893075d53a82a30c13855cbaa791402695b01e369f/hf_transfer-0.1.9-cp38-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:d2fde99d502093ade3ab1b53f80da18480e9902aa960dab7f74fb1b9e5bc5746" },
{ url = "https://mirrors.aliyun.com/pypi/packages/09/89/d4e234727a26b2546c8fb70a276cd924260d60135f2165bf8b9ed67bb9a4/hf_transfer-0.1.9-cp38-abi3-win32.whl", hash = "sha256:435cc3cdc8524ce57b074032b8fd76eed70a4224d2091232fa6a8cef8fd6803e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a1/14/f1e15b851d1c2af5b0b1a82bf8eb10bda2da62d98180220ba6fd8879bb5b/hf_transfer-0.1.9-cp38-abi3-win_amd64.whl", hash = "sha256:16f208fc678911c37e11aa7b586bc66a37d02e636208f18b6bc53d29b5df40ad" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
[[package]]
name = "hf-xet"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.2.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/5e/6e/0f11bacf08a67f7fb5ee09740f2ca54163863b07b70d579356e9222ce5d8/hf_xet-1.2.0.tar.gz", hash = "sha256:a8c27070ca547293b6890c4bf389f713f80e8c478631432962bb7f4bc0bd7d7f" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/9e/a5/85ef910a0aa034a2abcfadc360ab5ac6f6bc4e9112349bd40ca97551cff0/hf_xet-1.2.0-cp313-cp313t-macosx_10_12_x86_64.whl", hash = "sha256:ceeefcd1b7aed4956ae8499e2199607765fbd1c60510752003b6cc0b8413b649" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ea/40/e2e0a7eb9a51fe8828ba2d47fe22a7e74914ea8a0db68a18c3aa7449c767/hf_xet-1.2.0-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:b70218dd548e9840224df5638fdc94bd033552963cfa97f9170829381179c813" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a5/7d/daf7f8bc4594fdd59a8a596f9e3886133fdc68e675292218a5e4c1b7e834/hf_xet-1.2.0-cp313-cp313t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7d40b18769bb9a8bc82a9ede575ce1a44c75eb80e7375a01d76259089529b5dc" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b1/ba/45ea2f605fbf6d81c8b21e4d970b168b18a53515923010c312c06cd83164/hf_xet-1.2.0-cp313-cp313t-manylinux_2_28_aarch64.whl", hash = "sha256:cd3a6027d59cfb60177c12d6424e31f4b5ff13d8e3a1247b3a584bf8977e6df5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4a/1d/04513e3cab8f29ab8c109d309ddd21a2705afab9d52f2ba1151e0c14f086/hf_xet-1.2.0-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:6de1fc44f58f6dd937956c8d304d8c2dea264c80680bcfa61ca4a15e7b76780f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f0/7c/60a2756d7feec7387db3a1176c632357632fbe7849fce576c5559d4520c7/hf_xet-1.2.0-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:f182f264ed2acd566c514e45da9f2119110e48a87a327ca271027904c70c5832" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4e/64/48fffbd67fb418ab07451e4ce641a70de1c40c10a13e25325e24858ebe5a/hf_xet-1.2.0-cp313-cp313t-win_amd64.whl", hash = "sha256:293a7a3787e5c95d7be1857358a9130694a9c6021de3f27fa233f37267174382" },
{ url = "https://mirrors.aliyun.com/pypi/packages/96/2d/22338486473df5923a9ab7107d375dbef9173c338ebef5098ef593d2b560/hf_xet-1.2.0-cp37-abi3-macosx_10_12_x86_64.whl", hash = "sha256:46740d4ac024a7ca9b22bebf77460ff43332868b661186a8e46c227fdae01848" },
{ url = "https://mirrors.aliyun.com/pypi/packages/7f/8c/c5becfa53234299bc2210ba314eaaae36c2875e0045809b82e40a9544f0c/hf_xet-1.2.0-cp37-abi3-macosx_11_0_arm64.whl", hash = "sha256:27df617a076420d8845bea087f59303da8be17ed7ec0cd7ee3b9b9f579dff0e4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9a/92/cf3ab0b652b082e66876d08da57fcc6fa2f0e6c70dfbbafbd470bb73eb47/hf_xet-1.2.0-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3651fd5bfe0281951b988c0facbe726aa5e347b103a675f49a3fa8144c7968fd" },
{ url = "https://mirrors.aliyun.com/pypi/packages/46/92/3f7ec4a1b6a65bf45b059b6d4a5d38988f63e193056de2f420137e3c3244/hf_xet-1.2.0-cp37-abi3-manylinux_2_28_aarch64.whl", hash = "sha256:d06fa97c8562fb3ee7a378dd9b51e343bc5bc8190254202c9771029152f5e08c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0b/dd/7ac658d54b9fb7999a0ccb07ad863b413cbaf5cf172f48ebcd9497ec7263/hf_xet-1.2.0-cp37-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:4c1428c9ae73ec0939410ec73023c4f842927f39db09b063b9482dac5a3bb737" },
{ url = "https://mirrors.aliyun.com/pypi/packages/92/68/89ac4e5b12a9ff6286a12174c8538a5930e2ed662091dd2572bbe0a18c8a/hf_xet-1.2.0-cp37-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:a55558084c16b09b5ed32ab9ed38421e2d87cf3f1f89815764d1177081b99865" },
{ url = "https://mirrors.aliyun.com/pypi/packages/cb/44/870d44b30e1dcfb6a65932e3e1506c103a8a5aea9103c337e7a53180322c/hf_xet-1.2.0-cp37-abi3-win_amd64.whl", hash = "sha256:e6584a52253f72c9f52f9e549d5895ca7a471608495c4ecaa6cc73dba2b24d69" },
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "hpack"
Removed onnxruntime (#4632) ### What problem does this PR solve? Removed onnxruntime. It conflicts with the onnxruntime-gpu. ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-01-24 23:41:52 +08:00
version = "4.1.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/2c/48/71de9ed269fdae9c8057e5a4c0aa7402e8bb16f2c6e90b3aa53327b113f8/hpack-4.1.0.tar.gz", hash = "sha256:ec5eca154f7056aa06f196a557655c5b009b382873ac8d1e66e79e87535f1dca" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/07/c6/80c95b1b2b94682a72cbdbfb85b81ae2daffa4291fbfa1b1464502ede10d/hpack-4.1.0-py3-none-any.whl", hash = "sha256:157ac792668d995c657d93111f46b4535ed114f0c9c8d672271bbec7eae1b496" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "html-text"
version = "0.6.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "lxml" },
{ name = "lxml-html-clean" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/5e/35/10ab103fec3b953ed1ba56ff827d00afc170dd204e0785c5fe7d3d6e1ae9/html_text-0.6.2.tar.gz", hash = "sha256:81455b4de5430cf63ce7c45a870fb8629e79ca8518e240f172d62409c2f2ff72" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/b8/bd/b28e8456b952268083642d631bec7d74e1564a54bfd2e6f6d13e597bbec0/html_text-0.6.2-py2.py3-none-any.whl", hash = "sha256:d83d619ccd4b4d6172e21084d8a46e29e49ce87a08cc02161e7ca8c2918e7bca" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "html5lib"
version = "1.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "six" },
{ name = "webencodings" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ac/b6/b55c3f49042f1df3dcd422b7f224f939892ee94f22abcf503a9b7339eaf2/html5lib-1.1.tar.gz", hash = "sha256:b2e5b40261e20f354d198eae92afc10d750afb487ed5e50f9c4eaf07c184146f" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/6c/dd/a834df6482147d48e225a49515aabc28974ad5a4ca3215c18a882565b028/html5lib-1.1-py2.py3-none-any.whl", hash = "sha256:0d78f8fde1c230e99fe37986a60526d7049ed4bf8a9fadbad5f00e22e58e041d" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "httpcore"
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
version = "1.0.9"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "certifi" },
{ name = "h11" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/06/94/82699a10bca87a5556c9c59b5963f2d039dbd239f25bc2a63907a05a14cb/httpcore-1.0.9.tar.gz", hash = "sha256:6e34463af53fd2ab5d807f399a9b45ea31c3dfa2276f15a2c3f00afff6e176e8" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/7e/f5/f66802a942d491edb555dd61e3a9961140fd64c90bce1eafd741609d334d/httpcore-1.0.9-py3-none-any.whl", hash = "sha256:2d400746a40668fc9dec9810239072b40b4484b640a8c38fd654a024c7a1bf55" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "httplib2"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "0.31.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "pyparsing" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/52/77/6653db69c1f7ecfe5e3f9726fdadc981794656fcd7d98c4209fecfea9993/httplib2-0.31.0.tar.gz", hash = "sha256:ac7ab497c50975147d4f7b1ade44becc7df2f8954d42b38b3d69c515f531135c" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/8c/a2/0d269db0f6163be503775dc8b6a6fa15820cc9fdc866f6ba608d86b721f2/httplib2-0.31.0-py3-none-any.whl", hash = "sha256:b9cd78abea9b4e43a7714c6e0f8b6b8561a6fc1e95d5dbd367f5bf0ef35f5d24" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "httpx"
Fix problem with Google Cloud models with reasoning (like gemini) - Additional fix to issue #10474 (#10502) ### What problem does this PR solve? Issue #10474 - Update to PR #10477 ### Type of change - [X] Bug Fix (non-breaking change which fixes an issue)
2025-10-15 08:54:20 +02:00
version = "0.28.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "anyio" },
{ name = "certifi" },
{ name = "httpcore" },
{ name = "idna" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/b1/df/48c586a5fe32a0f01324ee087459e112ebb7224f646c0b5023f5e79e9956/httpx-0.28.1.tar.gz", hash = "sha256:75e98c5f16b0f35b567856f597f06ff2270a374470a5c2392242528e3e3e42fc" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/2a/39/e50c7c3a983047577ee07d2a9e53faf5a69493943ec3f6a384bdc792deb2/httpx-0.28.1-py3-none-any.whl", hash = "sha256:d909fcccc110f8c7faf814ca82a9a4d816bc5a6dbfea25d6591d6985b8ba59ad" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
[package.optional-dependencies]
http2 = [
{ name = "h2" },
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "httpx-sse"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "0.4.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/0f/4c/751061ffa58615a32c31b2d82e8482be8dd4a89154f003147acee90f2be9/httpx_sse-0.4.3.tar.gz", hash = "sha256:9b1ed0127459a66014aec3c56bebd93da3c1bc8bb6618c8082039a44889a755d" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/d2/fd/6668e5aec43ab844de6fc74927e155a3b37bf40d7c3790e49fc0406b6578/httpx_sse-0.4.3-py3-none-any.whl", hash = "sha256:0ac1c9fe3c0afad2e0ebb25a934a59f4c7823b60792691f779fad2c5568830fc" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "huggingface-hub"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.3.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "filelock" },
{ name = "fsspec" },
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
{ name = "hf-xet", marker = "platform_machine == 'AMD64' or platform_machine == 'aarch64' or platform_machine == 'amd64' or platform_machine == 'arm64' or platform_machine == 'x86_64'" },
{ name = "httpx" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "packaging" },
{ name = "pyyaml" },
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
{ name = "shellingham" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "tqdm" },
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
{ name = "typer-slim" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "typing-extensions" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/dd/dd/1cc985c5dda36298b152f75e82a1c81f52243b78fb7e9cad637a29561ad1/huggingface_hub-1.3.1.tar.gz", hash = "sha256:e80e0cfb4a75557c51ab20d575bdea6bb6106c2f97b7c75d8490642f1efb6df5" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/90/fb/cb8fe5f71d5622427f20bcab9e06a696a5aaf21bfe7bd0a8a0c63c88abf5/huggingface_hub-1.3.1-py3-none-any.whl", hash = "sha256:efbc7f3153cb84e2bb69b62ed90985e21ecc9343d15647a419fc0ee4b85f0ac3" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "humanfriendly"
version = "10.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "pyreadline3", marker = "sys_platform == 'win32'" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/cc/3f/2c29224acb2e2df4d2046e4c73ee2662023c58ff5b113c4c1adac0886c43/humanfriendly-10.0.tar.gz", hash = "sha256:6b0b831ce8f15f7300721aa49829fc4e83921a9a301cc7f606be6686a2288ddc" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/f0/0f/310fb31e39e2d734ccaa2c0fb981ee41f7bd5056ce9bc29b2248bd569169/humanfriendly-10.0-py2.py3-none-any.whl", hash = "sha256:1697e1a8a8f550fd43c2865cd84542fc175a61dcb779b6fee18cf6b6ccba1477" },
Feat: Alter flask to Quart for async API serving. (#11275) ### What problem does this PR solve? #11277 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-18 17:05:16 +08:00
]
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
[[package]]
name = "humanize"
version = "4.15.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ba/66/a3921783d54be8a6870ac4ccffcd15c4dc0dd7fcce51c6d63b8c63935276/humanize-4.15.0.tar.gz", hash = "sha256:1dd098483eb1c7ee8e32eb2e99ad1910baefa4b75c3aff3a82f4d78688993b10" }
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/c5/7b/bca5613a0c3b542420cf92bd5e5fb8ebd5435ce1011a091f66bb7693285e/humanize-4.15.0-py3-none-any.whl", hash = "sha256:b1186eb9f5a9749cd9cb8565aee77919dd7c8d076161cf44d70e59e3301e1769" },
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
]
Feat: Alter flask to Quart for async API serving. (#11275) ### What problem does this PR solve? #11277 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-18 17:05:16 +08:00
[[package]]
name = "hypercorn"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "0.18.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Alter flask to Quart for async API serving. (#11275) ### What problem does this PR solve? #11277 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-18 17:05:16 +08:00
dependencies = [
{ name = "h11" },
{ name = "h2" },
{ name = "priority" },
{ name = "wsproto" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/44/01/39f41a014b83dd5c795217362f2ca9071cf243e6a75bdcd6cd5b944658cc/hypercorn-0.18.0.tar.gz", hash = "sha256:d63267548939c46b0247dc8e5b45a9947590e35e64ee73a23c074aa3cf88e9da" }
Feat: Alter flask to Quart for async API serving. (#11275) ### What problem does this PR solve? #11277 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-18 17:05:16 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/93/35/850277d1b17b206bd10874c8a9a3f52e059452fb49bb0d22cbb908f6038b/hypercorn-0.18.0-py3-none-any.whl", hash = "sha256:225e268f2c1c2f28f6d8f6db8f40cb8c992963610c5725e13ccfcddccb24b1cd" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "hyperframe"
Removed onnxruntime (#4632) ### What problem does this PR solve? Removed onnxruntime. It conflicts with the onnxruntime-gpu. ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-01-24 23:41:52 +08:00
version = "6.1.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/02/e7/94f8232d4a74cc99514c13a9f995811485a6903d48e5d952771ef6322e30/hyperframe-6.1.0.tar.gz", hash = "sha256:f630908a00854a7adeabd6382b43923a4c4cd4b821fcb527e6ab9e15382a3b08" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/48/30/47d0bf6072f7252e6521f3447ccfa40b421b6824517f82854703d0f5a98b/hyperframe-6.1.0-py3-none-any.whl", hash = "sha256:b03380493a519fce58ea5af42e4a42317bf9bd425596f7a0835ffce80f1a42e5" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: sync test group to top pyproject.toml (#8015) ### What problem does this PR solve? sync test group from sdk/python/pyproject.toml to top pyproject.toml ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-06-03 15:21:06 +08:00
[[package]]
name = "hypothesis"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "6.150.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: sync test group to top pyproject.toml (#8015) ### What problem does this PR solve? sync test group from sdk/python/pyproject.toml to top pyproject.toml ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-06-03 15:21:06 +08:00
dependencies = [
{ name = "sortedcontainers" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ad/4e/cd3a398b9834386a79f4eb777dc4004ca439c1019d324771ec8196fc8354/hypothesis-6.150.1.tar.gz", hash = "sha256:dc79672b3771e92e6563ca0c56a24135438f319b257a1a1982deb8fbb791be89" }
Feat: sync test group to top pyproject.toml (#8015) ### What problem does this PR solve? sync test group from sdk/python/pyproject.toml to top pyproject.toml ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-06-03 15:21:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/22/18/f43815244cd99b54d8ac9f44f9799bb7c0115e48e29bc7a1899c0589ee48/hypothesis-6.150.1-py3-none-any.whl", hash = "sha256:7badb28a0da323d6afaf25eae1c93932cb8ac06193355f5e080d6e6465a51da5" },
Feat: sync test group to top pyproject.toml (#8015) ### What problem does this PR solve? sync test group from sdk/python/pyproject.toml to top pyproject.toml ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-06-03 15:21:06 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "hyppo"
Bump beartype to 0.22.6 (#11463) ### What problem does this PR solve? Bump beartype to 0.22.6 ### Type of change - [x] Refactoring
2025-11-22 11:56:43 +08:00
version = "0.5.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "autograd" },
Bump beartype to 0.22.6 (#11463) ### What problem does this PR solve? Bump beartype to 0.22.6 ### Type of change - [x] Refactoring
2025-11-22 11:56:43 +08:00
{ name = "future" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "numba" },
{ name = "numpy" },
Bump beartype to 0.22.6 (#11463) ### What problem does this PR solve? Bump beartype to 0.22.6 ### Type of change - [x] Refactoring
2025-11-22 11:56:43 +08:00
{ name = "pandas" },
{ name = "patsy" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "scikit-learn" },
Bump infinity to v0.6.11. Requires python>=3.11 (#11814) ### What problem does this PR solve? Bump infinity to v0.6.11. Requires python>=3.11 ### Type of change - [x] Refactoring
2025-12-09 16:23:37 +08:00
{ name = "scipy" },
Bump beartype to 0.22.6 (#11463) ### What problem does this PR solve? Bump beartype to 0.22.6 ### Type of change - [x] Refactoring
2025-11-22 11:56:43 +08:00
{ name = "statsmodels" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/dd/a6/0d84fe8486a1447da8bdb8ebb249d525fd8c1d0fe038bceb003c6e0513f9/hyppo-0.5.2.tar.gz", hash = "sha256:4634d15516248a43d25c241ed18beeb79bb3210360f7253693b3f154fe8c9879" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ea/c4/d46858cfac3c0aad314a1fc378beae5c8cac499b677650a34b5a6a3d4328/hyppo-0.5.2-py3-none-any.whl", hash = "sha256:5cc18f9e158fe2cf1804c9a1e979e807118ee89a303f29dc5cb8891d92d44ef3" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "idna"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "3.11"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/6f/6d/0703ccc57f3a7233505399edb88de3cbd678da106337b9fcde432b65ed60/idna-3.11.tar.gz", hash = "sha256:795dafcc9c04ed0c1fb032c2aa73654d8e8c5023a7df64a53f39190ada629902" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/0e/61/66938bbb5fc52dbdf84594873d5b51fb1f7c7794e9c0f5bd885f30bc507b/idna-3.11-py3-none-any.whl", hash = "sha256:771a87f49d9defaf64091e6e6fe9c18d4833f140bd19464795bc32d966ca37ea" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "ijson"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "3.4.0.post0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/2d/30/7ab4b9e88e7946f6beef419f74edcc541df3ea562c7882257b4eaa82417d/ijson-3.4.0.post0.tar.gz", hash = "sha256:9aa02dc70bb245670a6ca7fba737b992aeeb4895360980622f7e568dbf23e41e" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/1b/20/aaec6977f9d538bbadd760c7fa0f6a0937742abdcc920ec6478a8576e55f/ijson-3.4.0.post0-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:114ed248166ac06377e87a245a158d6b98019d2bdd3bb93995718e0bd996154f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/5b/29/06bf56a866e2fe21453a1ad8f3a5d7bca3c723f73d96329656dfee969783/ijson-3.4.0.post0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:ffb21203736b08fe27cb30df6a4f802fafb9ef7646c5ff7ef79569b63ea76c57" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ba/ae/e1d0fda91ba7a444b75f0d60cb845fdb1f55d3111351529dcbf4b1c276fe/ijson-3.4.0.post0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:07f20ecd748602ac7f18c617637e53bd73ded7f3b22260bba3abe401a7fc284e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4d/24/5a24533be2726396cc1724dc237bada09b19715b5bfb0e7b9400db0901ad/ijson-3.4.0.post0-cp313-cp313-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:27aa193d47ffc6bc4e45453896ad98fb089a367e8283b973f1fe5c0198b60b4e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/05/60/026c3efcec23c329657e878cbc0a9a25b42e7eb3971e8c2377cb3284e2b7/ijson-3.4.0.post0-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:ccddb2894eb7af162ba43b9475ac5825d15d568832f82eb8783036e5d2aebd42" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ed/c2/036499909b7a1bc0bcd85305e4348ad171aeb9df57581287533bdb3497e9/ijson-3.4.0.post0-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:61ab0b8c5bf707201dc67e02c116f4b6545c4afd7feb2264b989d242d9c4348a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ba/75/e7736073ad96867c129f9e799e3e65086badd89dbf3911f76d9b3bf8a115/ijson-3.4.0.post0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:254cfb8c124af68327a0e7a49b50bbdacafd87c4690a3d62c96eb01020a685ef" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9d/1b/1c1575d2cda136985561fcf774fe6c54412cd0fa08005342015af0403193/ijson-3.4.0.post0-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:04ac9ca54db20f82aeda6379b5f4f6112fdb150d09ebce04affeab98a17b4ed3" },
{ url = "https://mirrors.aliyun.com/pypi/packages/28/4d/aba9871feb624df8494435d1a9ddc7b6a4f782c6044bfc0d770a4b59f145/ijson-3.4.0.post0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:a603d7474bf35e7b3a8e49c8dabfc4751841931301adff3f3318171c4e407f32" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3f/9a/791baa83895fb6e492bce2c7a0ea6427b6a41fe854349e62a37d0c9deaf0/ijson-3.4.0.post0-cp313-cp313-win32.whl", hash = "sha256:ec5bb1520cb212ebead7dba048bb9b70552c3440584f83b01b0abc96862e2a09" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a9/0c/061f51493e1da21116d74ee8f6a6b9ae06ca5fa2eb53c3b38b64f9a9a5ae/ijson-3.4.0.post0-cp313-cp313-win_amd64.whl", hash = "sha256:3505dff18bdeb8b171eb28af6df34857e2be80dc01e2e3b624e77215ad58897f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c7/89/4344e176f2c5f5ef3251c9bfa4ddd5b4cf3f9601fd6ec3f677a3ba0b9c71/ijson-3.4.0.post0-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:45a0b1c833ed2620eaf8da958f06ac8351c59e5e470e078400d23814670ed708" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d4/b1/85012c586a6645f9fb8bfa3ef62ed2f303c8d73fc7c2f705111582925980/ijson-3.4.0.post0-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:7809ec8c8f40228edaaa089f33e811dff4c5b8509702652870d3f286c9682e27" },
{ url = "https://mirrors.aliyun.com/pypi/packages/65/ea/7b7e2815c101d78b33e74d64ddb70cccc377afccd5dda76e566ed3fcb56f/ijson-3.4.0.post0-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:cf4a34c2cfe852aee75c89c05b0a4531c49dc0be27eeed221afd6fbf9c3e149c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/59/7d/2175e599cb77a64f528629bad3ce95dfdf2aa6171d313c1fc00bbfaf0d22/ijson-3.4.0.post0-cp313-cp313t-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:a39d5d36067604b26b78de70b8951c90e9272450642661fe531a8f7a6936a7fa" },
{ url = "https://mirrors.aliyun.com/pypi/packages/13/97/82247c501c92405bb2fc44ab5efb497335bcb9cf0f5d3a0b04a800737bd8/ijson-3.4.0.post0-cp313-cp313t-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:83fc738d81c9ea686b452996110b8a6678296c481e0546857db24785bff8da92" },
{ url = "https://mirrors.aliyun.com/pypi/packages/95/ca/b956f507bb02e05ce109fd11ab6a2c054f8b686cc5affe41afe50630984d/ijson-3.4.0.post0-cp313-cp313t-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:b2a81aee91633868f5b40280e2523f7c5392e920a5082f47c5e991e516b483f6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3e/12/e827840ab81d86a9882e499097934df53294f05155f1acfcb9a211ac1142/ijson-3.4.0.post0-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:56169e298c5a2e7196aaa55da78ddc2415876a74fe6304f81b1eb0d3273346f7" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1b/3b/59238d9422c31a4aefa22ebeb8e599e706158a0ab03669ef623be77a499a/ijson-3.4.0.post0-cp313-cp313t-musllinux_1_2_i686.whl", hash = "sha256:eeb9540f0b1a575cbb5968166706946458f98c16e7accc6f2fe71efa29864241" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b6/0f/ec01c36c128c37edb8a5ae8f3de3256009f886338d459210dfe121ee4ba9/ijson-3.4.0.post0-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:ba3478ff0bb49d7ba88783f491a99b6e3fa929c930ab062d2bb7837e6a38fe88" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c8/cf/5560e1db96c6d10a5313be76bf5a1754266cbfb5cc13ff64d107829e07b1/ijson-3.4.0.post0-cp313-cp313t-win32.whl", hash = "sha256:b005ce84e82f28b00bf777a464833465dfe3efa43a0a26c77b5ac40723e1a728" },
{ url = "https://mirrors.aliyun.com/pypi/packages/22/5a/cbb69144c3b25dd56f5421ff7dc0cf3051355579062024772518e4f4b3c5/ijson-3.4.0.post0-cp313-cp313t-win_amd64.whl", hash = "sha256:fe9c84c9b1c8798afa407be1cea1603401d99bfc7c34497e19f4f5e5ddc9b441" },
Feat:new api /sequence2txt and update QWenSeq2txt (#11643) ### What problem does this PR solve? change: new api /sequence2txt, update QWenSeq2txt and ZhipuSeq2txt ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-02 11:17:31 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "imagesize"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.4.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a7/84/62473fb57d61e31fef6e36d64a179c8781605429fd927b5dd608c997be31/imagesize-1.4.1.tar.gz", hash = "sha256:69150444affb9cb0d5cc5a92b3676f0b2fb7cd9ae39e947a5e11a36b4497cd4a" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ff/62/85c4c919272577931d407be5ba5d71c20f0b616d31a0befe0ae45bb79abd/imagesize-1.4.1-py2.py3-none-any.whl", hash = "sha256:0d8d18d08f840c19d0ee7ca1fd82490fdc3729b7ac93f49870406ddde8ef8d8b" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "importlib-metadata"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "8.7.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "zipp" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f3/49/3b30cad09e7771a4982d9975a8cbf64f00d4a1ececb53297f1d9a7be1b10/importlib_metadata-8.7.1.tar.gz", hash = "sha256:49fef1ae6440c182052f407c8d34a68f72efc36db9ca90dc0113398f2fdde8bb" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/fa/5e/f8e9a1d23b9c20a551a8a02ea3637b4642e22c2626e3a13a9a29cdea99eb/importlib_metadata-8.7.1-py3-none-any.whl", hash = "sha256:5a1f80bf1daa489495071efbb095d75a634cf28a8bc299581244063b53176151" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "infinity-emb"
version = "0.0.66"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "hf-transfer" },
{ name = "huggingface-hub" },
{ name = "numpy" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/25/f7/aa95638ba1aa5cc4ecdc9ac62a8db8715bf7975e17b4471cb86c7c7c1f56/infinity_emb-0.0.66.tar.gz", hash = "sha256:9c9a361ccebf8e8f626c1f685286518d03d0c35e7d14179ae7c2500b4fc68b98" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/6b/3a/a16d04f2af44295d60d4263455a136682816d03acd85a2f190051cc70288/infinity_emb-0.0.66-py3-none-any.whl", hash = "sha256:1dc6ed9fa48e6cbe83650a7583dbbb4bc393900c39c326bb0aff2ddc090ac018" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "infinity-sdk"
Bump to infinity v0.7.0 (#14968) ### What problem does this PR solve? Upgrade infinity ### Type of change - [x] Refactoring
2026-05-18 10:25:59 +08:00
version = "0.7.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
{ name = "datrie" },
{ name = "hanziconv" },
{ name = "nltk" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "numpy" },
{ name = "pandas" },
Bump infinity to v0.6.0-dev2 (#4497) ### What problem does this PR solve? Bump infinity to v0.6.0-dev2. Close #4477 ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-01-16 10:07:17 +08:00
{ name = "polars-lts-cpu" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "pyarrow" },
{ name = "pydantic" },
{ name = "readerwriterlock" },
{ name = "requests" },
{ name = "setuptools" },
bump infinity (#10422) ### What problem does this PR solve? bump infinity to v0.6.0-dev7 Needs https://github.com/infiniflow/infinity/pull/3016 ### Type of change - [x] Other (please describe): Infinity
2025-10-10 12:41:45 +08:00
{ name = "sqlglot", extra = ["rs"] },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "thrift" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/46/c9/0bb4981acdaca79b864b26d998201f652ef6024fce7d0c39da9c11605890/infinity_sdk-0.7.0.tar.gz", hash = "sha256:42ba8c6acd4fad918b1ed189fab3383023e4750d46ef1bf1c9465ffcf3ff8335" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/27/56/01a0b4b816c70595a83ad2d7ec387d1d991f0ea4607d77799bac010be27c/infinity_sdk-0.7.0-py3-none-any.whl", hash = "sha256:4772fded64ff733eb3dff36df3a2c1c867049acf3c6e54d4e40b5a915c0f3c18" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: add Airtable connector and integration for data synchronization (#12211) ### What problem does this PR solve? change: add Airtable connector and integration for data synchronization ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-25 17:50:41 +08:00
[[package]]
name = "inflection"
version = "0.5.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/e1/7e/691d061b7329bc8d54edbf0ec22fbfb2afe61facb681f9aaa9bff7a27d04/inflection-0.5.1.tar.gz", hash = "sha256:1a29730d366e996aaacffb2f1f1cb9593dc38e2ddd30c91250c6dde09ea9b417" }
Feat: add Airtable connector and integration for data synchronization (#12211) ### What problem does this PR solve? change: add Airtable connector and integration for data synchronization ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-25 17:50:41 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/59/91/aa6bde563e0085a02a435aa99b49ef75b0a4b062635e606dab23ce18d720/inflection-0.5.1-py2.py3-none-any.whl", hash = "sha256:f38b2b640938a4f35ade69ac3d053042959b62a0f1076a5bbaa1b9526605a8a2" },
Feat: add Airtable connector and integration for data synchronization (#12211) ### What problem does this PR solve? change: add Airtable connector and integration for data synchronization ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-25 17:50:41 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "iniconfig"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "2.3.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/72/34/14ca021ce8e5dfedc35312d08ba8bf51fdd999c576889fc2c24cb97f4f10/iniconfig-2.3.0.tar.gz", hash = "sha256:c76315c77db068650d49c5b56314774a7804df16fee4402c1f19d6d15d8c4730" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/cb/b1/3846dd7f199d53cb17f49cba7e651e9ce294d8497c8c150530ed11865bb8/iniconfig-2.3.0-py3-none-any.whl", hash = "sha256:f631c04d2c48c52b84d0d0549c99ff3859c98df65b3101406327ecc7d53fbf12" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
[[package]]
name = "inquirerpy"
version = "0.3.4"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pfzy" },
{ name = "prompt-toolkit" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/64/73/7570847b9da026e07053da3bbe2ac7ea6cde6bb2cbd3c7a5a950fa0ae40b/InquirerPy-0.3.4.tar.gz", hash = "sha256:89d2ada0111f337483cb41ae31073108b2ec1e618a49d7110b0d7ade89fc197e" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ce/ff/3b59672c47c6284e8005b42e84ceba13864aa0f39f067c973d1af02f5d91/InquirerPy-0.3.4-py3-none-any.whl", hash = "sha256:c65fdfbac1fa00e3ee4fb10679f4d3ed7a012abf4833910e63c295827fe2a7d4" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "inscriptis"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "2.7.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "lxml" },
{ name = "requests" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/5d/34/4124dc3dc52738ecf6e3fcb5a6671269e99e8bcbf1eadfb5b356c3b85174/inscriptis-2.7.0.tar.gz", hash = "sha256:52ee95e63611ba46481f0be5cf56988d4a1b9672e382c9b1cea2e0ff90bb29f3" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/3d/e3/08458a4bed3f04ee1ca16809b6c45907198c587b3b18dd3d37ac18cd180b/inscriptis-2.7.0-py3-none-any.whl", hash = "sha256:db368f67e7c0624df2fdff7bee1c3a74e795ff536fabce252e3ff29f9c28c23e" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: add local & ssh provider in admin panel (#15039) ### What problem does this PR solve? Feat: add local & ssh provider in admin panel ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-20 16:56:20 +08:00
[[package]]
name = "invoke"
version = "3.0.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/33/f6/227c48c5fe47fa178ccf1fda8f047d16c97ba926567b661e9ce2045c600c/invoke-3.0.3.tar.gz", hash = "sha256:437b6a622223824380bfb4e64f612711a6b648c795f565efc8625af66fb57f0c" }
Feat: add local & ssh provider in admin panel (#15039) ### What problem does this PR solve? Feat: add local & ssh provider in admin panel ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-20 16:56:20 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/5a/de/bbc12563bbf979618d17625a4e753ff7a078523e28d870d3626daa97261a/invoke-3.0.3-py3-none-any.whl", hash = "sha256:f11327165e5cbb89b2ad1d88d3292b5113332c43b8553b494da435d6ec6f5053" },
Feat: add local & ssh provider in admin panel (#15039) ### What problem does this PR solve? Feat: add local & ssh provider in admin panel ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-20 16:56:20 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "ir-datasets"
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
version = "0.5.11"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "beautifulsoup4" },
{ name = "ijson" },
{ name = "inscriptis" },
{ name = "lxml" },
{ name = "lz4" },
{ name = "numpy" },
Docs: prepare docs for release v0.17.1 (#5900) ### What problem does this PR solve? ### Type of change - [x] Documentation Update
2025-03-11 14:39:41 +08:00
{ name = "pyarrow" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "pyyaml" },
{ name = "requests" },
{ name = "tqdm" },
{ name = "trec-car-tools" },
{ name = "unlzw3" },
{ name = "warc3-wet" },
{ name = "warc3-wet-clueweb09" },
{ name = "zlib-state" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/80/c6/f02811c51fec845ee87a10bb3675516a2d71935b203e5ddb80b7eb59b1da/ir_datasets-0.5.11.tar.gz", hash = "sha256:06c90af634ae5063c813286b35065debca1a974d26e136403d899f3ecd7ad463" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/81/38/73fa582d6997362d9b4901b2a8ba1177b2d2896aa59ab8d069a3884e2e0d/ir_datasets-0.5.11-py3-none-any.whl", hash = "sha256:ae78549e5a7fa45e50462b7acb9f0765fc344fec6054108bf3dd063050555206" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "isodate"
version = "0.7.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/54/4d/e940025e2ce31a8ce1202635910747e5a87cc3a6a6bb2d00973375014749/isodate-0.7.2.tar.gz", hash = "sha256:4cd1aa0f43ca76f4a6c6c0292a85f40b35ec2e43e315b59f06e6d32171a953e6" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/15/aa/0aca39a37d3c7eb941ba736ede56d689e7be91cab5d9ca846bde3999eba6/isodate-0.7.2-py3-none-any.whl", hash = "sha256:28009937d8031054830160fce6d409ed342816b543597cece116d966c6d99e15" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "itsdangerous"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "2.2.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/9c/cb/8ac0172223afbccb63986cc25049b154ecfb5e85932587206f42317be31d/itsdangerous-2.2.0.tar.gz", hash = "sha256:e0050c0b7da1eea53ffaf149c0cfbb5c6e2e2b69c4bef22c81fa6eb73e5f6173" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/04/96/92447566d16df59b2a776c0fb82dbc4d9e07cd95062562af01e408583fc4/itsdangerous-2.2.0-py3-none-any.whl", hash = "sha256:c6242fc49e35958c8b15141343aa660db5fc54d4f13a1db01a3f5891b98700ef" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "jinja2"
Docs: prepare docs for release v0.17.1 (#5900) ### What problem does this PR solve? ### Type of change - [x] Documentation Update
2025-03-11 14:39:41 +08:00
version = "3.1.6"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "markupsafe" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/df/bf/f7da0350254c0ed7c72f3e33cef02e048281fec7ecec5f032d4aac52226b/jinja2-3.1.6.tar.gz", hash = "sha256:0137fb05990d35f1275a587e9aee6d56da821fc83491a0fb838183be43f66d6d" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/62/a1/3d680cbfd5f4b8f15abc1d571870c5fc3e594bb582bc3b64ea099db13e56/jinja2-3.1.6-py3-none-any.whl", hash = "sha256:85ece4451f492d0c13c5dd7c13a64681a86afae63a5f347908daf103ce6d2f67" },
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
]
[[package]]
name = "jira"
version = "3.10.5"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
dependencies = [
{ name = "defusedxml" },
{ name = "packaging" },
{ name = "requests" },
{ name = "requests-oauthlib" },
{ name = "requests-toolbelt" },
{ name = "typing-extensions" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/65/73/ee4daa7cf4eea457180de0ea78b730b44bb5ad2829dae49cf708a1460819/jira-3.10.5.tar.gz", hash = "sha256:2d09ae3bf4741a2787dd889dfea5926a5d509aac3b28ab3b98c098709e6ee72d" }
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/89/57/ad078d7379e390798559446607e413fc953c7510711462ab34194dba5924/jira-3.10.5-py3-none-any.whl", hash = "sha256:d4da1385c924ee693d6cc9838e56a34e31d74f0d6899934ef35bbd0d2d33997f" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "jiter"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "0.12.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/45/9d/e0660989c1370e25848bb4c52d061c71837239738ad937e83edca174c273/jiter-0.12.0.tar.gz", hash = "sha256:64dfcd7d5c168b38d3f9f8bba7fc639edb3418abcc74f22fdbe6b8938293f30b" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/3d/a6/97209693b177716e22576ee1161674d1d58029eb178e01866a0422b69224/jiter-0.12.0-cp313-cp313-macosx_10_12_x86_64.whl", hash = "sha256:6cc49d5130a14b732e0612bc76ae8db3b49898732223ef8b7599aa8d9810683e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/06/4d/125c5c1537c7d8ee73ad3d530a442d6c619714b95027143f1b61c0b4dfe0/jiter-0.12.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:37f27a32ce36364d2fa4f7fdc507279db604d27d239ea2e044c8f148410defe1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/99/bf/a840b89847885064c41a5f52de6e312e91fa84a520848ee56c97e4fa0205/jiter-0.12.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:bbc0944aa3d4b4773e348cda635252824a78f4ba44328e042ef1ff3f6080d1cf" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8a/88/e63441c28e0db50e305ae23e19c1d8fae012d78ed55365da392c1f34b09c/jiter-0.12.0-cp313-cp313-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:da25c62d4ee1ffbacb97fac6dfe4dcd6759ebdc9015991e92a6eae5816287f44" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0a/7c/49b02714af4343970eb8aca63396bc1c82fa01197dbb1e9b0d274b550d4e/jiter-0.12.0-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:048485c654b838140b007390b8182ba9774621103bd4d77c9c3f6f117474ba45" },
{ url = "https://mirrors.aliyun.com/pypi/packages/69/ba/0a809817fdd5a1db80490b9150645f3aae16afad166960bcd562be194f3b/jiter-0.12.0-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:635e737fbb7315bef0037c19b88b799143d2d7d3507e61a76751025226b3ac87" },
{ url = "https://mirrors.aliyun.com/pypi/packages/5f/c3/c9fc0232e736c8877d9e6d83d6eeb0ba4e90c6c073835cc2e8f73fdeef51/jiter-0.12.0-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:4e017c417b1ebda911bd13b1e40612704b1f5420e30695112efdbed8a4b389ed" },
{ url = "https://mirrors.aliyun.com/pypi/packages/96/61/61f69b7e442e97ca6cd53086ddc1cf59fb830549bc72c0a293713a60c525/jiter-0.12.0-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:89b0bfb8b2bf2351fba36bb211ef8bfceba73ef58e7f0c68fb67b5a2795ca2f9" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e9/2e/76bb3332f28550c8f1eba3bf6e5efe211efda0ddbbaf24976bc7078d42a5/jiter-0.12.0-cp313-cp313-musllinux_1_1_aarch64.whl", hash = "sha256:f5aa5427a629a824a543672778c9ce0c5e556550d1569bb6ea28a85015287626" },
{ url = "https://mirrors.aliyun.com/pypi/packages/84/d6/fa96efa87dc8bff2094fb947f51f66368fa56d8d4fc9e77b25d7fbb23375/jiter-0.12.0-cp313-cp313-musllinux_1_1_x86_64.whl", hash = "sha256:ed53b3d6acbcb0fd0b90f20c7cb3b24c357fe82a3518934d4edfa8c6898e498c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8a/28/93f67fdb4d5904a708119a6ab58a8f1ec226ff10a94a282e0215402a8462/jiter-0.12.0-cp313-cp313-win32.whl", hash = "sha256:4747de73d6b8c78f2e253a2787930f4fffc68da7fa319739f57437f95963c4de" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c4/1f/30b0eb087045a0abe2a5c9c0c0c8da110875a1d3be83afd4a9a4e548be3c/jiter-0.12.0-cp313-cp313-win_amd64.whl", hash = "sha256:e25012eb0c456fcc13354255d0338cd5397cce26c77b2832b3c4e2e255ea5d9a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2c/f4/2b4daf99b96bce6fc47971890b14b2a36aef88d7beb9f057fafa032c6141/jiter-0.12.0-cp313-cp313-win_arm64.whl", hash = "sha256:c97b92c54fe6110138c872add030a1f99aea2401ddcdaa21edf74705a646dd60" },
{ url = "https://mirrors.aliyun.com/pypi/packages/39/ca/67bb15a7061d6fe20b9b2a2fd783e296a1e0f93468252c093481a2f00efa/jiter-0.12.0-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:53839b35a38f56b8be26a7851a48b89bc47e5d88e900929df10ed93b95fea3d6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/18/af/1788031cd22e29c3b14bc6ca80b16a39a0b10e611367ffd480c06a259831/jiter-0.12.0-cp313-cp313t-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:94f669548e55c91ab47fef8bddd9c954dab1938644e715ea49d7e117015110a4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/05/17/710bf8472d1dff0d3caf4ced6031060091c1320f84ee7d5dcbed1f352417/jiter-0.12.0-cp313-cp313t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:351d54f2b09a41600ffea43d081522d792e81dcfb915f6d2d242744c1cc48beb" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fb/f1/1dcc4618b59761fef92d10bcbb0b038b5160be653b003651566a185f1a5c/jiter-0.12.0-cp313-cp313t-win_amd64.whl", hash = "sha256:2a5e90604620f94bf62264e7c2c038704d38217b7465b863896c6d7c902b06c7" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d9/32/63cb1d9f1c5c6632a783c0052cde9ef7ba82688f7065e2f0d5f10a7e3edb/jiter-0.12.0-cp313-cp313t-win_arm64.whl", hash = "sha256:88ef757017e78d2860f96250f9393b7b577b06a956ad102c29c8237554380db3" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "jmespath"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
version = "0.10.0"
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/3c/56/3f325b1eef9791759784aa5046a8f6a1aff8f7c898a2e34506771d3b99d8/jmespath-0.10.0.tar.gz", hash = "sha256:b85d0567b8666149a93172712e68920734333c0ce7e89b78b3e987f71e5ed4f9" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/07/cb/5f001272b6faeb23c1c9e0acc04d48eaaf5c862c17709d20e3469c6e0139/jmespath-0.10.0-py2.py3-none-any.whl", hash = "sha256:cdf6525904cc597730141d61b36f2e4b8ecc257c420fa2f4549bac2c2d0cb72f" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "joblib"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "1.5.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/41/f2/d34e8b3a08a9cc79a50b2208a93dce981fe615b64d5a4d4abee421d898df/joblib-1.5.3.tar.gz", hash = "sha256:8561a3269e6801106863fd0d6d84bb737be9e7631e33aaed3fb9ce5953688da3" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/7b/91/984aca2ec129e2757d1e4e3c81c3fcda9d0f85b74670a094cc443d9ee949/joblib-1.5.3-py3-none-any.whl", hash = "sha256:5fc3c5039fc5ca8c0276333a188bbd59d6b7ab37fe6632daa76bc7f9ec18e713" },
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
]
[[package]]
name = "joserfc"
version = "1.7.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
dependencies = [
{ name = "cryptography" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/44/90/25cb27518750218e4f850be63d8bbb2343efaad1c01c3571aaa4b3c33bd7/joserfc-1.7.1.tar.gz", hash = "sha256:77d0b76514879c68c6f433bc5b7357a4ab72008ff1e33d8379fd11d72bd8ca81" }
Light GraphRAG (#4585) ### What problem does this PR solve? #4543 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-01-22 19:43:14 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/b3/00/fa62404c3e347f946faa13aa21085205f9cc06ad17671e37f81a51662ae8/joserfc-1.7.1-py3-none-any.whl", hash = "sha256:b3e3d655612e2e1ef67b2600f2f420e12e537b020208fab1761fad647319c164" },
Light GraphRAG (#4585) ### What problem does this PR solve? #4543 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-01-22 19:43:14 +08:00
]
[[package]]
name = "json-repair"
version = "0.35.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/95/9c/5ef83a13541c3444e0b949e88b3aa0f4e364e37acf4ffa9de476d36a3de0/json_repair-0.35.0.tar.gz", hash = "sha256:e70f834865a4ae5fe64352c23c1c16d3b70c5dd62dc544a169d8b0932bdbdcaa" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/4e/c7/9a63ff79f2f6350c0f5e504bad940386015381b4d171bd73077465a7dbbc/json_repair-0.35.0-py3-none-any.whl", hash = "sha256:1d429407158474d28a996e745b8f8f7dc78957cb2cfbc92120b9f580b5230a9e" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "jsonpath"
version = "0.82.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/cf/a1/693351acd0a9edca4de9153372a65e75398898ea7f8a5c722ab00f464929/jsonpath-0.82.2.tar.gz", hash = "sha256:d87ef2bcbcded68ee96bc34c1809b69457ecec9b0c4dd471658a12bd391002d1" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "jsonschema"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "4.26.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "attrs" },
{ name = "jsonschema-specifications" },
{ name = "referencing" },
{ name = "rpds-py" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/b3/fc/e067678238fa451312d4c62bf6e6cf5ec56375422aee02f9cb5f909b3047/jsonschema-4.26.0.tar.gz", hash = "sha256:0c26707e2efad8aa1bfc5b7ce170f3fccc2e4918ff85989ba9ffa9facb2be326" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/69/90/f63fb5873511e014207a475e2bb4e8b2e570d655b00ac19a9a0ca0a385ee/jsonschema-4.26.0-py3-none-any.whl", hash = "sha256:d489f15263b8d200f8387e64b4c3a75f06629559fb73deb8fdfb525f2dab50ce" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "jsonschema-specifications"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "2025.9.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "referencing" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/19/74/a633ee74eb36c44aa6d1095e7cc5569bebf04342ee146178e2d36600708b/jsonschema_specifications-2025.9.1.tar.gz", hash = "sha256:b540987f239e745613c7a9176f3edb72b832a4ac465cf02712288397832b5e8d" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/41/45/1a4ed80516f02155c51f51e8cedb3c1902296743db0bbc66608a0db2814f/jsonschema_specifications-2025.9.1-py3-none-any.whl", hash = "sha256:98802fee3a11ee76ecaca44429fda8a41bff98b00a0f2838151b113f210cc6fe" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "kaitaistruct"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "0.11"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/27/b8/ca7319556912f68832daa4b81425314857ec08dfccd8dbc8c0f65c992108/kaitaistruct-0.11.tar.gz", hash = "sha256:053ee764288e78b8e53acf748e9733268acbd579b8d82a427b1805453625d74b" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/4a/4a/cf14bf3b1f5ffb13c69cf5f0ea78031247790558ee88984a8bdd22fae60d/kaitaistruct-0.11-py2.py3-none-any.whl", hash = "sha256:5c6ce79177b4e193a577ecd359e26516d1d6d000a0bffd6e1010f2a46a62a561" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "kiwisolver"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.4.9"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/5c/3c/85844f1b0feb11ee581ac23fe5fce65cd049a200c1446708cc1b7f922875/kiwisolver-1.4.9.tar.gz", hash = "sha256:c3b22c26c6fd6811b0ae8363b95ca8ce4ea3c202d3d0975b2914310ceb1bcc4d" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/31/c1/c2686cda909742ab66c7388e9a1a8521a59eb89f8bcfbee28fc980d07e24/kiwisolver-1.4.9-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:a5d0432ccf1c7ab14f9949eec60c5d1f924f17c037e9f8b33352fa05799359b8" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ca/f0/f44f50c9f5b1a1860261092e3bc91ecdc9acda848a8b8c6abfda4a24dd5c/kiwisolver-1.4.9-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:efb3a45b35622bb6c16dbfab491a8f5a391fe0e9d45ef32f4df85658232ca0e2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2d/7a/9d90a151f558e29c3936b8a47ac770235f436f2120aca41a6d5f3d62ae8d/kiwisolver-1.4.9-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:1a12cf6398e8a0a001a059747a1cbf24705e18fe413bc22de7b3d15c67cffe3f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e9/e9/f218a2cb3a9ffbe324ca29a9e399fa2d2866d7f348ec3a88df87fc248fc5/kiwisolver-1.4.9-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:b67e6efbf68e077dd71d1a6b37e43e1a99d0bff1a3d51867d45ee8908b931098" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d9/28/aac26d4c882f14de59041636292bc838db8961373825df23b8eeb807e198/kiwisolver-1.4.9-cp313-cp313-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:5656aa670507437af0207645273ccdfee4f14bacd7f7c67a4306d0dcaeaf6eed" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8b/ad/8bfc1c93d4cc565e5069162f610ba2f48ff39b7de4b5b8d93f69f30c4bed/kiwisolver-1.4.9-cp313-cp313-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:bfc08add558155345129c7803b3671cf195e6a56e7a12f3dde7c57d9b417f525" },
{ url = "https://mirrors.aliyun.com/pypi/packages/da/f1/6aca55ff798901d8ce403206d00e033191f63d82dd708a186e0ed2067e9c/kiwisolver-1.4.9-cp313-cp313-manylinux_2_24_s390x.manylinux_2_28_s390x.whl", hash = "sha256:40092754720b174e6ccf9e845d0d8c7d8e12c3d71e7fc35f55f3813e96376f78" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d1/91/eed031876c595c81d90d0f6fc681ece250e14bf6998c3d7c419466b523b7/kiwisolver-1.4.9-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:497d05f29a1300d14e02e6441cf0f5ee81c1ff5a304b0d9fb77423974684e08b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e9/ec/4d1925f2e49617b9cca9c34bfa11adefad49d00db038e692a559454dfb2e/kiwisolver-1.4.9-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:bdd1a81a1860476eb41ac4bc1e07b3f07259e6d55bbf739b79c8aaedcf512799" },
{ url = "https://mirrors.aliyun.com/pypi/packages/43/cb/450cd4499356f68802750c6ddc18647b8ea01ffa28f50d20598e0befe6e9/kiwisolver-1.4.9-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:e6b93f13371d341afee3be9f7c5964e3fe61d5fa30f6a30eb49856935dfe4fc3" },
{ url = "https://mirrors.aliyun.com/pypi/packages/71/67/fc76242bd99f885651128a5d4fa6083e5524694b7c88b489b1b55fdc491d/kiwisolver-1.4.9-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:d75aa530ccfaa593da12834b86a0724f58bff12706659baa9227c2ccaa06264c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/75/bd/f1a5d894000941739f2ae1b65a32892349423ad49c2e6d0771d0bad3fae4/kiwisolver-1.4.9-cp313-cp313-win_amd64.whl", hash = "sha256:dd0a578400839256df88c16abddf9ba14813ec5f21362e1fe65022e00c883d4d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/95/38/dce480814d25b99a391abbddadc78f7c117c6da34be68ca8b02d5848b424/kiwisolver-1.4.9-cp313-cp313-win_arm64.whl", hash = "sha256:d4188e73af84ca82468f09cadc5ac4db578109e52acb4518d8154698d3a87ca2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e2/37/7d218ce5d92dadc5ebdd9070d903e0c7cf7edfe03f179433ac4d13ce659c/kiwisolver-1.4.9-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:5a0f2724dfd4e3b3ac5a82436a8e6fd16baa7d507117e4279b660fe8ca38a3a1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/23/b0/e85a2b48233daef4b648fb657ebbb6f8367696a2d9548a00b4ee0eb67803/kiwisolver-1.4.9-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:1b11d6a633e4ed84fc0ddafd4ebfd8ea49b3f25082c04ad12b8315c11d504dc1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/44/98/f2425bc0113ad7de24da6bb4dae1343476e95e1d738be7c04d31a5d037fd/kiwisolver-1.4.9-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:61874cdb0a36016354853593cffc38e56fc9ca5aa97d2c05d3dcf6922cd55a11" },
{ url = "https://mirrors.aliyun.com/pypi/packages/98/d8/594657886df9f34c4177cc353cc28ca7e6e5eb562d37ccc233bff43bbe2a/kiwisolver-1.4.9-cp313-cp313t-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:60c439763a969a6af93b4881db0eed8fadf93ee98e18cbc35bc8da868d0c4f0c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/5c/c6/38a115b7170f8b306fc929e166340c24958347308ea3012c2b44e7e295db/kiwisolver-1.4.9-cp313-cp313t-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:92a2f997387a1b79a75e7803aa7ded2cfbe2823852ccf1ba3bcf613b62ae3197" },
{ url = "https://mirrors.aliyun.com/pypi/packages/bf/3b/e04883dace81f24a568bcee6eb3001da4ba05114afa622ec9b6fafdc1f5e/kiwisolver-1.4.9-cp313-cp313t-manylinux_2_24_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:a31d512c812daea6d8b3be3b2bfcbeb091dbb09177706569bcfc6240dcf8b41c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9f/80/20ace48e33408947af49d7d15c341eaee69e4e0304aab4b7660e234d6288/kiwisolver-1.4.9-cp313-cp313t-manylinux_2_24_s390x.manylinux_2_28_s390x.whl", hash = "sha256:52a15b0f35dad39862d376df10c5230155243a2c1a436e39eb55623ccbd68185" },
{ url = "https://mirrors.aliyun.com/pypi/packages/64/31/6ce4380a4cd1f515bdda976a1e90e547ccd47b67a1546d63884463c92ca9/kiwisolver-1.4.9-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:a30fd6fdef1430fd9e1ba7b3398b5ee4e2887783917a687d86ba69985fb08748" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fa/e9/3f3fcba3bcc7432c795b82646306e822f3fd74df0ee81f0fa067a1f95668/kiwisolver-1.4.9-cp313-cp313t-musllinux_1_2_ppc64le.whl", hash = "sha256:cc9617b46837c6468197b5945e196ee9ca43057bb7d9d1ae688101e4e1dddf64" },
{ url = "https://mirrors.aliyun.com/pypi/packages/99/43/7320c50e4133575c66e9f7dadead35ab22d7c012a3b09bb35647792b2a6d/kiwisolver-1.4.9-cp313-cp313t-musllinux_1_2_s390x.whl", hash = "sha256:0ab74e19f6a2b027ea4f845a78827969af45ce790e6cb3e1ebab71bdf9f215ff" },
{ url = "https://mirrors.aliyun.com/pypi/packages/65/d6/17ae4a270d4a987ef8a385b906d2bdfc9fce502d6dc0d3aea865b47f548c/kiwisolver-1.4.9-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:dba5ee5d3981160c28d5490f0d1b7ed730c22470ff7f6cc26cfcfaacb9896a07" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2a/8f/8f6f491d595a9e5912971f3f863d81baddccc8a4d0c3749d6a0dd9ffc9df/kiwisolver-1.4.9-cp313-cp313t-win_arm64.whl", hash = "sha256:0749fd8f4218ad2e851e11cc4dc05c7cbc0cbc4267bdfdb31782e65aace4ee9c" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: add basic Langfuse support for LLM module (#6443) ### What problem does this PR solve? #6155 Add basic Langfuse support for LLM module. A trace example: <img width="755" alt="image" src="https://github.com/user-attachments/assets/25c1f852-5116-486c-a47f-6097187142ca" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-03-24 13:18:47 +08:00
[[package]]
name = "langfuse"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "4.7.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: add basic Langfuse support for LLM module (#6443) ### What problem does this PR solve? #6155 Add basic Langfuse support for LLM module. A trace example: <img width="755" alt="image" src="https://github.com/user-attachments/assets/25c1f852-5116-486c-a47f-6097187142ca" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-03-24 13:18:47 +08:00
dependencies = [
{ name = "backoff" },
{ name = "httpx" },
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
{ name = "opentelemetry-api" },
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
{ name = "opentelemetry-exporter-otlp-proto-http" },
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
{ name = "opentelemetry-sdk" },
Feat: add basic Langfuse support for LLM module (#6443) ### What problem does this PR solve? #6155 Add basic Langfuse support for LLM module. A trace example: <img width="755" alt="image" src="https://github.com/user-attachments/assets/25c1f852-5116-486c-a47f-6097187142ca" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-03-24 13:18:47 +08:00
{ name = "packaging" },
{ name = "pydantic" },
{ name = "wrapt" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/2c/74/a6f1a99893ee6d1a69439ae7eb92f8fe8806103492dc26531d5942dbd3bf/langfuse-4.7.1.tar.gz", hash = "sha256:f9e262eceedb353b191c1da1f8452d1e8ebf52297ca20e160cda0206608e3a40" }
Feat: add basic Langfuse support for LLM module (#6443) ### What problem does this PR solve? #6155 Add basic Langfuse support for LLM module. A trace example: <img width="755" alt="image" src="https://github.com/user-attachments/assets/25c1f852-5116-486c-a47f-6097187142ca" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-03-24 13:18:47 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/9f/9a/bd3368f46b6c72ee2068b80536826b02ae86df53eff1c79941344503098f/langfuse-4.7.1-py3-none-any.whl", hash = "sha256:a4e59c81ad5e5b16a65d3849f4923ebc3ad6e67ec803ada83d50c0cb66149490" },
Feat: add basic Langfuse support for LLM module (#6443) ### What problem does this PR solve? #6155 Add basic Langfuse support for LLM module. A trace example: <img width="755" alt="image" src="https://github.com/user-attachments/assets/25c1f852-5116-486c-a47f-6097187142ca" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-03-24 13:18:47 +08:00
]
Feat: add admin CLI and admin service (#10186) ### What problem does this PR solve? Introduce new feature: RAGFlow system admin service and CLI ### Introduction Admin Service is a dedicated management component designed to monitor, maintain, and administrate the RAGFlow system. It provides comprehensive tools for ensuring system stability, performing operational tasks, and managing users and permissions efficiently. The service offers monitoring of critical components, including the RAGFlow server, Task Executor processes, and dependent services such as MySQL, Infinity / Elasticsearch, Redis, and MinIO. It automatically checks their health status, resource usage, and uptime, and performs restarts in case of failures to minimize downtime. For user and system management, it supports listing, creating, modifying, and deleting users and their associated resources like knowledge bases and Agents. Built with scalability and reliability in mind, the Admin Service ensures smooth system operation and simplifies maintenance workflows. It consists of a server-side Service and a command-line client (CLI), both implemented in Python. User commands are parsed using the Lark parsing toolkit. - **Admin Service**: A backend service that interfaces with the RAGFlow system to execute administrative operations and monitor its status. - **Admin CLI**: A command-line interface that allows users to connect to the Admin Service and issue commands for system management. ### Starting the Admin Service 1. Before start Admin Service, please make sure RAGFlow system is already started. 2. Run the service script: ```bash python admin/admin_server.py ``` The service will start and listen for incoming connections from the CLI on the configured port. ### Using the Admin CLI 1. Ensure the Admin Service is running. 2. Launch the CLI client: ```bash python admin/admin_client.py -h 0.0.0.0 -p 9381 ## Supported Commands Commands are case-insensitive and must be terminated with a semicolon (`;`). ### Service Management Commands - [x] `LIST SERVICES;` - Lists all available services within the RAGFlow system. - [ ] `SHOW SERVICE <id>;` - Shows detailed status information for the service identified by `<id>`. - [ ] `STARTUP SERVICE <id>;` - Attempts to start the service identified by `<id>`. - [ ] `SHUTDOWN SERVICE <id>;` - Attempts to gracefully shut down the service identified by `<id>`. - [ ] `RESTART SERVICE <id>;` - Attempts to restart the service identified by `<id>`. ### User Management Commands - [x] `LIST USERS;` - Lists all users known to the system. - [ ] `SHOW USER '<username>';` - Shows details and permissions for the specified user. The username must be enclosed in single or double quotes. - [ ] `DROP USER '<username>';` - Removes the specified user from the system. Use with caution. - [ ] `ALTER USER PASSWORD '<username>' '<new_password>';` - Changes the password for the specified user. ### Data and Agent Commands - [ ] `LIST DATASETS OF '<username>';` - Lists the datasets associated with the specified user. - [ ] `LIST AGENTS OF '<username>';` - Lists the agents associated with the specified user. ### Meta-Commands Meta-commands are prefixed with a backslash (`\`). - `\?` or `\help` - Shows help information for the available commands. - `\q` or `\quit` - Exits the CLI application. ## Examples ```commandline admin> list users; +-------------------------------+------------------------+-----------+-------------+ | create_date | email | is_active | nickname | +-------------------------------+------------------------+-----------+-------------+ | Fri, 22 Nov 2024 16:03:41 GMT | jeffery@infiniflow.org | 1 | Jeffery | | Fri, 22 Nov 2024 16:10:55 GMT | aya@infiniflow.org | 1 | Waterdancer | +-------------------------------+------------------------+-----------+-------------+ admin> list services; +-------------------------------------------------------------------------------------------+-----------+----+---------------+-------+----------------+ | extra | host | id | name | port | service_type | +-------------------------------------------------------------------------------------------+-----------+----+---------------+-------+----------------+ | {} | 0.0.0.0 | 0 | ragflow_0 | 9380 | ragflow_server | | {'meta_type': 'mysql', 'password': 'infini_rag_flow', 'username': 'root'} | localhost | 1 | mysql | 5455 | meta_data | | {'password': 'infini_rag_flow', 'store_type': 'minio', 'user': 'rag_flow'} | localhost | 2 | minio | 9000 | file_store | | {'password': 'infini_rag_flow', 'retrieval_type': 'elasticsearch', 'username': 'elastic'} | localhost | 3 | elasticsearch | 1200 | retrieval | | {'db_name': 'default_db', 'retrieval_type': 'infinity'} | localhost | 4 | infinity | 23817 | retrieval | | {'database': 1, 'mq_type': 'redis', 'password': 'infini_rag_flow'} | localhost | 5 | redis | 6379 | message_queue | +-------------------------------------------------------------------------------------------+-----------+----+---------------+-------+----------------+ ``` ### Type of change - [x] New Feature (non-breaking change which adds functionality) Signed-off-by: jinhai <haijin.chn@gmail.com>
2025-09-22 10:37:49 +08:00
[[package]]
name = "lark"
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
version = "1.3.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/da/34/28fff3ab31ccff1fd4f6c7c7b0ceb2b6968d8ea4950663eadcb5720591a0/lark-1.3.1.tar.gz", hash = "sha256:b426a7a6d6d53189d318f2b6236ab5d6429eaf09259f1ca33eb716eed10d2905" }
Feat: add admin CLI and admin service (#10186) ### What problem does this PR solve? Introduce new feature: RAGFlow system admin service and CLI ### Introduction Admin Service is a dedicated management component designed to monitor, maintain, and administrate the RAGFlow system. It provides comprehensive tools for ensuring system stability, performing operational tasks, and managing users and permissions efficiently. The service offers monitoring of critical components, including the RAGFlow server, Task Executor processes, and dependent services such as MySQL, Infinity / Elasticsearch, Redis, and MinIO. It automatically checks their health status, resource usage, and uptime, and performs restarts in case of failures to minimize downtime. For user and system management, it supports listing, creating, modifying, and deleting users and their associated resources like knowledge bases and Agents. Built with scalability and reliability in mind, the Admin Service ensures smooth system operation and simplifies maintenance workflows. It consists of a server-side Service and a command-line client (CLI), both implemented in Python. User commands are parsed using the Lark parsing toolkit. - **Admin Service**: A backend service that interfaces with the RAGFlow system to execute administrative operations and monitor its status. - **Admin CLI**: A command-line interface that allows users to connect to the Admin Service and issue commands for system management. ### Starting the Admin Service 1. Before start Admin Service, please make sure RAGFlow system is already started. 2. Run the service script: ```bash python admin/admin_server.py ``` The service will start and listen for incoming connections from the CLI on the configured port. ### Using the Admin CLI 1. Ensure the Admin Service is running. 2. Launch the CLI client: ```bash python admin/admin_client.py -h 0.0.0.0 -p 9381 ## Supported Commands Commands are case-insensitive and must be terminated with a semicolon (`;`). ### Service Management Commands - [x] `LIST SERVICES;` - Lists all available services within the RAGFlow system. - [ ] `SHOW SERVICE <id>;` - Shows detailed status information for the service identified by `<id>`. - [ ] `STARTUP SERVICE <id>;` - Attempts to start the service identified by `<id>`. - [ ] `SHUTDOWN SERVICE <id>;` - Attempts to gracefully shut down the service identified by `<id>`. - [ ] `RESTART SERVICE <id>;` - Attempts to restart the service identified by `<id>`. ### User Management Commands - [x] `LIST USERS;` - Lists all users known to the system. - [ ] `SHOW USER '<username>';` - Shows details and permissions for the specified user. The username must be enclosed in single or double quotes. - [ ] `DROP USER '<username>';` - Removes the specified user from the system. Use with caution. - [ ] `ALTER USER PASSWORD '<username>' '<new_password>';` - Changes the password for the specified user. ### Data and Agent Commands - [ ] `LIST DATASETS OF '<username>';` - Lists the datasets associated with the specified user. - [ ] `LIST AGENTS OF '<username>';` - Lists the agents associated with the specified user. ### Meta-Commands Meta-commands are prefixed with a backslash (`\`). - `\?` or `\help` - Shows help information for the available commands. - `\q` or `\quit` - Exits the CLI application. ## Examples ```commandline admin> list users; +-------------------------------+------------------------+-----------+-------------+ | create_date | email | is_active | nickname | +-------------------------------+------------------------+-----------+-------------+ | Fri, 22 Nov 2024 16:03:41 GMT | jeffery@infiniflow.org | 1 | Jeffery | | Fri, 22 Nov 2024 16:10:55 GMT | aya@infiniflow.org | 1 | Waterdancer | +-------------------------------+------------------------+-----------+-------------+ admin> list services; +-------------------------------------------------------------------------------------------+-----------+----+---------------+-------+----------------+ | extra | host | id | name | port | service_type | +-------------------------------------------------------------------------------------------+-----------+----+---------------+-------+----------------+ | {} | 0.0.0.0 | 0 | ragflow_0 | 9380 | ragflow_server | | {'meta_type': 'mysql', 'password': 'infini_rag_flow', 'username': 'root'} | localhost | 1 | mysql | 5455 | meta_data | | {'password': 'infini_rag_flow', 'store_type': 'minio', 'user': 'rag_flow'} | localhost | 2 | minio | 9000 | file_store | | {'password': 'infini_rag_flow', 'retrieval_type': 'elasticsearch', 'username': 'elastic'} | localhost | 3 | elasticsearch | 1200 | retrieval | | {'db_name': 'default_db', 'retrieval_type': 'infinity'} | localhost | 4 | infinity | 23817 | retrieval | | {'database': 1, 'mq_type': 'redis', 'password': 'infini_rag_flow'} | localhost | 5 | redis | 6379 | message_queue | +-------------------------------------------------------------------------------------------+-----------+----+---------------+-------+----------------+ ``` ### Type of change - [x] New Feature (non-breaking change which adds functionality) Signed-off-by: jinhai <haijin.chn@gmail.com>
2025-09-22 10:37:49 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/82/3d/14ce75ef66813643812f3093ab17e46d3a206942ce7376d31ec2d36229e7/lark-1.3.1-py3-none-any.whl", hash = "sha256:c629b661023a014c37da873b4ff58a817398d12635d3bbb2c5a03be7fe5d1e12" },
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
]
[[package]]
name = "lark-oapi"
version = "1.6.8"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
dependencies = [
{ name = "httpx" },
{ name = "pycryptodome" },
{ name = "requests" },
{ name = "requests-toolbelt" },
{ name = "websockets" },
]
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/4a/ad/1ab04db5d549ad1a7a2cd33682b1c38dee1d65019cb24fd7a23270e6337d/lark_oapi-1.6.8-py3-none-any.whl", hash = "sha256:9b443a5d47a7d204dd42dc40896c8b75087cc35788e45c48c140806d7df7e5e8" },
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
]
[[package]]
name = "line-bot-sdk"
version = "3.23.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
dependencies = [
{ name = "aenum" },
{ name = "aiohttp" },
{ name = "deprecated" },
{ name = "future" },
{ name = "pydantic" },
{ name = "python-dateutil" },
{ name = "requests" },
{ name = "urllib3" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/b3/07/d5cd69937689799d3e64825e20af2af384798278750132346da10ffd8e5f/line_bot_sdk-3.23.0.tar.gz", hash = "sha256:33380c43e488e346e698bf564a4d613cd6ff0518024168ccbfdc32e28f082a01" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/10/7d/416aa1c5e26ae9f78c57d28510952efaa23e97d8a6229354eeb6d0cce47c/line_bot_sdk-3.23.0-py2.py3-none-any.whl", hash = "sha256:29a0bc206f8cc6b965e5b44e9e946108252b0bbe698115af00aeece12df6cd10" },
Feat: add admin CLI and admin service (#10186) ### What problem does this PR solve? Introduce new feature: RAGFlow system admin service and CLI ### Introduction Admin Service is a dedicated management component designed to monitor, maintain, and administrate the RAGFlow system. It provides comprehensive tools for ensuring system stability, performing operational tasks, and managing users and permissions efficiently. The service offers monitoring of critical components, including the RAGFlow server, Task Executor processes, and dependent services such as MySQL, Infinity / Elasticsearch, Redis, and MinIO. It automatically checks their health status, resource usage, and uptime, and performs restarts in case of failures to minimize downtime. For user and system management, it supports listing, creating, modifying, and deleting users and their associated resources like knowledge bases and Agents. Built with scalability and reliability in mind, the Admin Service ensures smooth system operation and simplifies maintenance workflows. It consists of a server-side Service and a command-line client (CLI), both implemented in Python. User commands are parsed using the Lark parsing toolkit. - **Admin Service**: A backend service that interfaces with the RAGFlow system to execute administrative operations and monitor its status. - **Admin CLI**: A command-line interface that allows users to connect to the Admin Service and issue commands for system management. ### Starting the Admin Service 1. Before start Admin Service, please make sure RAGFlow system is already started. 2. Run the service script: ```bash python admin/admin_server.py ``` The service will start and listen for incoming connections from the CLI on the configured port. ### Using the Admin CLI 1. Ensure the Admin Service is running. 2. Launch the CLI client: ```bash python admin/admin_client.py -h 0.0.0.0 -p 9381 ## Supported Commands Commands are case-insensitive and must be terminated with a semicolon (`;`). ### Service Management Commands - [x] `LIST SERVICES;` - Lists all available services within the RAGFlow system. - [ ] `SHOW SERVICE <id>;` - Shows detailed status information for the service identified by `<id>`. - [ ] `STARTUP SERVICE <id>;` - Attempts to start the service identified by `<id>`. - [ ] `SHUTDOWN SERVICE <id>;` - Attempts to gracefully shut down the service identified by `<id>`. - [ ] `RESTART SERVICE <id>;` - Attempts to restart the service identified by `<id>`. ### User Management Commands - [x] `LIST USERS;` - Lists all users known to the system. - [ ] `SHOW USER '<username>';` - Shows details and permissions for the specified user. The username must be enclosed in single or double quotes. - [ ] `DROP USER '<username>';` - Removes the specified user from the system. Use with caution. - [ ] `ALTER USER PASSWORD '<username>' '<new_password>';` - Changes the password for the specified user. ### Data and Agent Commands - [ ] `LIST DATASETS OF '<username>';` - Lists the datasets associated with the specified user. - [ ] `LIST AGENTS OF '<username>';` - Lists the agents associated with the specified user. ### Meta-Commands Meta-commands are prefixed with a backslash (`\`). - `\?` or `\help` - Shows help information for the available commands. - `\q` or `\quit` - Exits the CLI application. ## Examples ```commandline admin> list users; +-------------------------------+------------------------+-----------+-------------+ | create_date | email | is_active | nickname | +-------------------------------+------------------------+-----------+-------------+ | Fri, 22 Nov 2024 16:03:41 GMT | jeffery@infiniflow.org | 1 | Jeffery | | Fri, 22 Nov 2024 16:10:55 GMT | aya@infiniflow.org | 1 | Waterdancer | +-------------------------------+------------------------+-----------+-------------+ admin> list services; +-------------------------------------------------------------------------------------------+-----------+----+---------------+-------+----------------+ | extra | host | id | name | port | service_type | +-------------------------------------------------------------------------------------------+-----------+----+---------------+-------+----------------+ | {} | 0.0.0.0 | 0 | ragflow_0 | 9380 | ragflow_server | | {'meta_type': 'mysql', 'password': 'infini_rag_flow', 'username': 'root'} | localhost | 1 | mysql | 5455 | meta_data | | {'password': 'infini_rag_flow', 'store_type': 'minio', 'user': 'rag_flow'} | localhost | 2 | minio | 9000 | file_store | | {'password': 'infini_rag_flow', 'retrieval_type': 'elasticsearch', 'username': 'elastic'} | localhost | 3 | elasticsearch | 1200 | retrieval | | {'db_name': 'default_db', 'retrieval_type': 'infinity'} | localhost | 4 | infinity | 23817 | retrieval | | {'database': 1, 'mq_type': 'redis', 'password': 'infini_rag_flow'} | localhost | 5 | redis | 6379 | message_queue | +-------------------------------------------------------------------------------------------+-----------+----+---------------+-------+----------------+ ``` ### Type of change - [x] New Feature (non-breaking change which adds functionality) Signed-off-by: jinhai <haijin.chn@gmail.com>
2025-09-22 10:37:49 +08:00
]
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
[[package]]
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
name = "litellm"
feat(agent): Go port — canvas engine, 22 components, DSL v2, 13 endpoints (#15952) Ports the agent canvas subsystem from Python to Go. ## What's included ### Canvas Engine (Phase 0/1) - State engine, scheduler, variable resolver, Redis checkpoint store, cancel protocol - **209 tests** across canvas / component / io packages ### 22 Components (P0–P4) | Tier | Components | |---|---| | P0 T1+T2+T3 | LLM, Agent, ExitLoop, Switch, Categorize, Begin, Message, Invoke | | P1 T3 | VariableAggregator, VariableAssigner, StringTransform, ListOperations, DataOperations | | P2 T3 | Iteration, IterationItem, Loop, LoopItem | | P3 T3 | UserFillUp, Fillup | | P4 T5 | Browser, ExcelProcessor, DocsGenerator | ### DSL v2 Schema (Phase 2.5) - Typed v2 in-memory model with v1-to-v2 auto-detect converter - v1 legacy field stripping per plan §2.11.7 ### HTTP Endpoints & Bug Fixes (Plans PR1–PR3) - **DELETE SQL bug fix**: gorm v2 `Where("id = ?", id).Delete(...)` pattern - **CreateAgent validation**: title/DSL required, duplicate check, 103 envelope - **13 new endpoints**: templates, prompts, tags, sessions CRUD, chat/completions (SSE + non-stream stubs), rerun, test_db_connection, logs, webhook/logs - **756 Go unit tests** (745 → 756, +18) - **17 → 0 Python integration test failures** (test_agents.py + test_session_management/) ### Tools 21 eino tools: HTTPHelper, search tools, financial/data tools, mandatory stubs ### Infrastructure OTel observability, NATS message queue, DeepDoc gRPC client, SSRF guards, IDOR mitigation
2026-06-12 22:58:28 +08:00
version = "1.82.5"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
{ name = "aiohttp" },
{ name = "click" },
{ name = "fastuuid" },
{ name = "httpx" },
{ name = "importlib-metadata" },
{ name = "jinja2" },
{ name = "jsonschema" },
{ name = "openai" },
{ name = "pydantic" },
{ name = "python-dotenv" },
{ name = "tiktoken" },
{ name = "tokenizers" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/d7/f0/ec42ee14b388ce1d08a1df638f894ed7f1e6ac35b9daf0588ff7f7d52262/litellm-1.82.5.tar.gz", hash = "sha256:7988a9b48c8ccd9e5ebced80a4dfce9ce87083b303c3f67082450a4ad6dd312f" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/68/1f/b6c8043eec81eade53a4d0e15a50b788ab0e82661e01a25e0b8536a4dca0/litellm-1.82.5-py3-none-any.whl", hash = "sha256:e1012ab816352215c4e00776dd48b0c68058b537888a8ff82cca62af19e6fb11" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "llvmlite"
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
version = "0.46.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/74/cd/08ae687ba099c7e3d21fe2ea536500563ef1943c5105bf6ab4ee3829f68e/llvmlite-0.46.0.tar.gz", hash = "sha256:227c9fd6d09dce2783c18b754b7cd9d9b3b3515210c46acc2d3c5badd9870ceb" }
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/e6/ff/3eba7eb0aed4b6fca37125387cd417e8c458e750621fce56d2c541f67fa8/llvmlite-0.46.0-cp313-cp313-macosx_12_0_arm64.whl", hash = "sha256:30b60892d034bc560e0ec6654737aaa74e5ca327bd8114d82136aa071d611172" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0e/54/737755c0a91558364b9200702c3c9c15d70ed63f9b98a2c32f1c2aa1f3ba/llvmlite-0.46.0-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:6cc19b051753368a9c9f31dc041299059ee91aceec81bd57b0e385e5d5bf1a54" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e6/91/14f32e1d70905c1c0aa4e6609ab5d705c3183116ca02ac6df2091868413a/llvmlite-0.46.0-cp313-cp313-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:bca185892908f9ede48c0acd547fe4dc1bafefb8a4967d47db6cf664f9332d12" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4a/a7/d526ae86708cea531935ae777b6dbcabe7db52718e6401e0fb9c5edea80e/llvmlite-0.46.0-cp313-cp313-win_amd64.whl", hash = "sha256:67438fd30e12349ebb054d86a5a1a57fd5e87d264d2451bcfafbbbaa25b82a35" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "lxml"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "5.4.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/76/3d/14e82fc7c8fb1b7761f7e748fd47e2ec8276d137b6acfe5a4bb73853e08f/lxml-5.4.0.tar.gz", hash = "sha256:d12832e1dbea4be280b22fd0ea7c9b87f0d8fc51ba06e92dc62d52f804f78ebd" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/87/cb/2ba1e9dd953415f58548506fa5549a7f373ae55e80c61c9041b7fd09a38a/lxml-5.4.0-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:773e27b62920199c6197130632c18fb7ead3257fce1ffb7d286912e56ddb79e0" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b5/3e/6602a4dca3ae344e8609914d6ab22e52ce42e3e1638c10967568c5c1450d/lxml-5.4.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:ce9c671845de9699904b1e9df95acfe8dfc183f2310f163cdaa91a3535af95de" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4c/72/bf00988477d3bb452bef9436e45aeea82bb40cdfb4684b83c967c53909c7/lxml-5.4.0-cp313-cp313-manylinux_2_12_i686.manylinux2010_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:9454b8d8200ec99a224df8854786262b1bd6461f4280064c807303c642c05e76" },
{ url = "https://mirrors.aliyun.com/pypi/packages/92/1f/93e42d93e9e7a44b2d3354c462cd784dbaaf350f7976b5d7c3f85d68d1b1/lxml-5.4.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:cccd007d5c95279e529c146d095f1d39ac05139de26c098166c4beb9374b0f4d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/45/0b/363009390d0b461cf9976a499e83b68f792e4c32ecef092f3f9ef9c4ba54/lxml-5.4.0-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:0fce1294a0497edb034cb416ad3e77ecc89b313cff7adbee5334e4dc0d11f422" },
{ url = "https://mirrors.aliyun.com/pypi/packages/19/dc/6056c332f9378ab476c88e301e6549a0454dbee8f0ae16847414f0eccb74/lxml-5.4.0-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:24974f774f3a78ac12b95e3a20ef0931795ff04dbb16db81a90c37f589819551" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ee/8a/f8c66bbb23ecb9048a46a5ef9b495fd23f7543df642dabeebcb2eeb66592/lxml-5.4.0-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:497cab4d8254c2a90bf988f162ace2ddbfdd806fce3bda3f581b9d24c852e03c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/04/57/2e537083c3f381f83d05d9b176f0d838a9e8961f7ed8ddce3f0217179ce3/lxml-5.4.0-cp313-cp313-manylinux_2_28_aarch64.whl", hash = "sha256:e794f698ae4c5084414efea0f5cc9f4ac562ec02d66e1484ff822ef97c2cadff" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d8/80/ea8c4072109a350848f1157ce83ccd9439601274035cd045ac31f47f3417/lxml-5.4.0-cp313-cp313-manylinux_2_28_ppc64le.whl", hash = "sha256:2c62891b1ea3094bb12097822b3d44b93fc6c325f2043c4d2736a8ff09e65f60" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b3/47/c4be287c48cdc304483457878a3f22999098b9a95f455e3c4bda7ec7fc72/lxml-5.4.0-cp313-cp313-manylinux_2_28_s390x.whl", hash = "sha256:142accb3e4d1edae4b392bd165a9abdee8a3c432a2cca193df995bc3886249c8" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2f/04/6ef935dc74e729932e39478e44d8cfe6a83550552eaa072b7c05f6f22488/lxml-5.4.0-cp313-cp313-manylinux_2_28_x86_64.whl", hash = "sha256:1a42b3a19346e5601d1b8296ff6ef3d76038058f311902edd574461e9c036982" },
{ url = "https://mirrors.aliyun.com/pypi/packages/cb/f9/c33fc8daa373ef8a7daddb53175289024512b6619bc9de36d77dca3df44b/lxml-5.4.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:4291d3c409a17febf817259cb37bc62cb7eb398bcc95c1356947e2871911ae61" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8d/30/fc92bb595bcb878311e01b418b57d13900f84c2b94f6eca9e5073ea756e6/lxml-5.4.0-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:4f5322cf38fe0e21c2d73901abf68e6329dc02a4994e483adbcf92b568a09a54" },
{ url = "https://mirrors.aliyun.com/pypi/packages/43/d1/3ba7bd978ce28bba8e3da2c2e9d5ae3f8f521ad3f0ca6ea4788d086ba00d/lxml-5.4.0-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:0be91891bdb06ebe65122aa6bf3fc94489960cf7e03033c6f83a90863b23c58b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ee/cd/95fa2201041a610c4d08ddaf31d43b98ecc4b1d74b1e7245b1abdab443cb/lxml-5.4.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:15a665ad90054a3d4f397bc40f73948d48e36e4c09f9bcffc7d90c87410e478a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2d/a6/31da006fead660b9512d08d23d31e93ad3477dd47cc42e3285f143443176/lxml-5.4.0-cp313-cp313-win32.whl", hash = "sha256:d5663bc1b471c79f5c833cffbc9b87d7bf13f87e055a5c86c363ccd2348d7e82" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fc/14/c115516c62a7d2499781d2d3d7215218c0731b2c940753bf9f9b7b73924d/lxml-5.4.0-cp313-cp313-win_amd64.whl", hash = "sha256:bcb7a1096b4b6b24ce1ac24d4942ad98f983cd3810f9711bcd0293f43a9d8b9f" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Bump python to >=3.12 (#11846) ### What problem does this PR solve? Bump python to >=3.12 ### Type of change - [x] Refactoring
2025-12-09 19:55:25 +08:00
[package.optional-dependencies]
html-clean = [
{ name = "lxml-html-clean" },
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "lxml-html-clean"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "0.4.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "lxml" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/d9/cb/c9c5bb2a9c47292e236a808dd233a03531f53b626f36259dcd32b49c76da/lxml_html_clean-0.4.3.tar.gz", hash = "sha256:c9df91925b00f836c807beab127aac82575110eacff54d0a75187914f1bd9d8c" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/10/4a/63a9540e3ca73709f4200564a737d63a4c8c9c4dd032bab8535f507c190a/lxml_html_clean-0.4.3-py3-none-any.whl", hash = "sha256:63fd7b0b9c3a2e4176611c2ca5d61c4c07ffca2de76c14059a81a2825833731e" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "lz4"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "4.4.5"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/57/51/f1b86d93029f418033dddf9b9f79c8d2641e7454080478ee2aab5123173e/lz4-4.4.5.tar.gz", hash = "sha256:5f0b9e53c1e82e88c10d7c180069363980136b9d7a8306c4dca4f760d60c39f0" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/2f/46/08fd8ef19b782f301d56a9ccfd7dafec5fd4fc1a9f017cf22a1accb585d7/lz4-4.4.5-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:6bb05416444fafea170b07181bc70640975ecc2a8c92b3b658c554119519716c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8f/3f/ea3334e59de30871d773963997ecdba96c4584c5f8007fd83cfc8f1ee935/lz4-4.4.5-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:b424df1076e40d4e884cfcc4c77d815368b7fb9ebcd7e634f937725cd9a8a72a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/41/7b/7b3a2a0feb998969f4793c650bb16eff5b06e80d1f7bff867feb332f2af2/lz4-4.4.5-cp313-cp313-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:216ca0c6c90719731c64f41cfbd6f27a736d7e50a10b70fad2a9c9b262ec923d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/89/d1/f1d259352227bb1c185288dd694121ea303e43404aa77560b879c90e7073/lz4-4.4.5-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:533298d208b58b651662dd972f52d807d48915176e5b032fb4f8c3b6f5fe535c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d2/fb/ba9256c48266a09012ed1d9b0253b9aa4fe9cdff094f8febf5b26a4aa2a2/lz4-4.4.5-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:451039b609b9a88a934800b5fc6ee401c89ad9c175abf2f4d9f8b2e4ef1afc64" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a5/6d/dee32a9430c8b0e01bbb4537573cabd00555827f1a0a42d4e24ca803935c/lz4-4.4.5-cp313-cp313-win32.whl", hash = "sha256:a5f197ffa6fc0e93207b0af71b302e0a2f6f29982e5de0fbda61606dd3a55832" },
{ url = "https://mirrors.aliyun.com/pypi/packages/18/e0/f06028aea741bbecb2a7e9648f4643235279a770c7ffaf70bd4860c73661/lz4-4.4.5-cp313-cp313-win_amd64.whl", hash = "sha256:da68497f78953017deb20edff0dba95641cc86e7423dfadf7c0264e1ac60dc22" },
{ url = "https://mirrors.aliyun.com/pypi/packages/61/72/5bef44afb303e56078676b9f2486f13173a3c1e7f17eaac1793538174817/lz4-4.4.5-cp313-cp313-win_arm64.whl", hash = "sha256:c1cfa663468a189dab510ab231aad030970593f997746d7a324d40104db0d0a9" },
{ url = "https://mirrors.aliyun.com/pypi/packages/49/55/6a5c2952971af73f15ed4ebfdd69774b454bd0dc905b289082ca8664fba1/lz4-4.4.5-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:67531da3b62f49c939e09d56492baf397175ff39926d0bd5bd2d191ac2bff95f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4e/d7/fd62cbdbdccc35341e83aabdb3f6d5c19be2687d0a4eaf6457ddf53bba64/lz4-4.4.5-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:a1acbbba9edbcbb982bc2cac5e7108f0f553aebac1040fbec67a011a45afa1ba" },
{ url = "https://mirrors.aliyun.com/pypi/packages/77/69/225ffadaacb4b0e0eb5fd263541edd938f16cd21fe1eae3cd6d5b6a259dc/lz4-4.4.5-cp313-cp313t-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:a482eecc0b7829c89b498fda883dbd50e98153a116de612ee7c111c8bcf82d1d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c6/9e/2ce59ba4a21ea5dc43460cba6f34584e187328019abc0e66698f2b66c881/lz4-4.4.5-cp313-cp313t-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:e099ddfaa88f59dd8d36c8a3c66bd982b4984edf127eb18e30bb49bdba68ce67" },
{ url = "https://mirrors.aliyun.com/pypi/packages/80/4f/4d946bd1624ec229b386a3bc8e7a85fa9a963d67d0a62043f0af0978d3da/lz4-4.4.5-cp313-cp313t-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:a2af2897333b421360fdcce895c6f6281dc3fab018d19d341cf64d043fc8d90d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/02/a2/d429ba4720a9064722698b4b754fb93e42e625f1318b8fe834086c7c783b/lz4-4.4.5-cp313-cp313t-win32.whl", hash = "sha256:66c5de72bf4988e1b284ebdd6524c4bead2c507a2d7f172201572bac6f593901" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4b/85/7ba10c9b97c06af6c8f7032ec942ff127558863df52d866019ce9d2425cf/lz4-4.4.5-cp313-cp313t-win_amd64.whl", hash = "sha256:cdd4bdcbaf35056086d910d219106f6a04e1ab0daa40ec0eeef1626c27d0fddb" },
{ url = "https://mirrors.aliyun.com/pypi/packages/77/4d/a175459fb29f909e13e57c8f475181ad8085d8d7869bd8ad99033e3ee5fa/lz4-4.4.5-cp313-cp313t-win_arm64.whl", hash = "sha256:28ccaeb7c5222454cd5f60fcd152564205bcb801bd80e125949d2dfbadc76bbd" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
[[package]]
name = "mammoth"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.11.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
dependencies = [
{ name = "cobble" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ed/3c/a58418d2af00f2da60d4a51e18cd0311307b72d48d2fffec36a97b4a5e44/mammoth-1.11.0.tar.gz", hash = "sha256:a0f59e442f34d5b6447f4b0999306cbf3e67aaabfa8cb516f878fb1456744637" }
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ca/54/2e39566a131b13f6d8d193f974cb6a34e81bb7cc2fa6f7e03de067b36588/mammoth-1.11.0-py2.py3-none-any.whl", hash = "sha256:c077ab0d450bd7c0c6ecd529a23bf7e0fa8190c929e28998308ff4eada3f063b" },
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "markdown"
version = "3.6"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/22/02/4785861427848cc11e452cc62bb541006a1087cf04a1de83aedd5530b948/Markdown-3.6.tar.gz", hash = "sha256:ed4f41f6daecbeeb96e576ce414c41d2d876daa9a16cb35fa8ed8c2ddfad0224" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/fc/b3/0c0c994fe49cd661084f8d5dc06562af53818cc0abefaca35bdc894577c3/Markdown-3.6-py3-none-any.whl", hash = "sha256:48f276f4d8cfb8ce6527c8f79e2ee29708508bf4d40aa410fbc3b4ee832c850f" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "markdown-it-py"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "4.0.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "mdurl" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/5b/f5/4ec618ed16cc4f8fb3b701563655a69816155e79e24a17b651541804721d/markdown_it_py-4.0.0.tar.gz", hash = "sha256:cb0a2b4aa34f932c007117b194e945bd74e0ec24133ceb5bac59009cda1cb9f3" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/94/54/e7d793b573f298e1c9013b8c4dade17d481164aa517d1d7148619c2cedbf/markdown_it_py-4.0.0-py3-none-any.whl", hash = "sha256:87327c59b172c5011896038353a81343b6754500a08cd7a4973bb48c6d578147" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "markdown-to-json"
version = "2.1.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/b5/1a/d235321eac5ba6de9f83dd172b9549eb03fd149ecda4c8c25cdc9a5224bc/markdown_to_json-2.1.1.tar.gz", hash = "sha256:27642c42acd9130d1449f791f57fd0c4bbf58c7a76cfb5af6d42010ca97b1107" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/47/2b/dac4143951a16c0c03e8fe217c9fa784838d02a29c52ef0e8b265befea8f/markdown_to_json-2.1.1-py3-none-any.whl", hash = "sha256:c73b8a3ac7fbde65463dbaeba8bb925d1d54377cbb01a064cd65e1f3e394bd62" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
[[package]]
name = "markdownify"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "1.2.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
dependencies = [
{ name = "beautifulsoup4" },
{ name = "six" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/3f/bc/c8c8eea5335341306b0fa7e1cb33c5e1c8d24ef70ddd684da65f41c49c92/markdownify-1.2.2.tar.gz", hash = "sha256:b274f1b5943180b031b699b199cbaeb1e2ac938b75851849a31fd0c3d6603d09" }
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/43/ce/f1e3e9d959db134cedf06825fae8d5b294bd368aacdd0831a3975b7c4d55/markdownify-1.2.2-py3-none-any.whl", hash = "sha256:3f02d3cc52714084d6e589f70397b6fc9f2f3a8531481bf35e8cc39f975e186a" },
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "markupsafe"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "3.0.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/7e/99/7690b6d4034fffd95959cbe0c02de8deb3098cc577c67bb6a24fe5d7caa7/markupsafe-3.0.3.tar.gz", hash = "sha256:722695808f4b6457b320fdc131280796bdceb04ab50fe1795cd540799ebe1698" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/38/2f/907b9c7bbba283e68f20259574b13d005c121a0fa4c175f9bed27c4597ff/markupsafe-3.0.3-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:e1cf1972137e83c5d4c136c43ced9ac51d0e124706ee1c8aa8532c1287fa8795" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9c/d9/5f7756922cdd676869eca1c4e3c0cd0df60ed30199ffd775e319089cb3ed/markupsafe-3.0.3-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:116bb52f642a37c115f517494ea5feb03889e04df47eeff5b130b1808ce7c219" },
{ url = "https://mirrors.aliyun.com/pypi/packages/00/07/575a68c754943058c78f30db02ee03a64b3c638586fba6a6dd56830b30a3/markupsafe-3.0.3-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:133a43e73a802c5562be9bbcd03d090aa5a1fe899db609c29e8c8d815c5f6de6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a9/21/9b05698b46f218fc0e118e1f8168395c65c8a2c750ae2bab54fc4bd4e0e8/markupsafe-3.0.3-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:ccfcd093f13f0f0b7fdd0f198b90053bf7b2f02a3927a30e63f3ccc9df56b676" },
{ url = "https://mirrors.aliyun.com/pypi/packages/7f/71/544260864f893f18b6827315b988c146b559391e6e7e8f7252839b1b846a/markupsafe-3.0.3-cp313-cp313-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:509fa21c6deb7a7a273d629cf5ec029bc209d1a51178615ddf718f5918992ab9" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c2/28/b50fc2f74d1ad761af2f5dcce7492648b983d00a65b8c0e0cb457c82ebbe/markupsafe-3.0.3-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:a4afe79fb3de0b7097d81da19090f4df4f8d3a2b3adaa8764138aac2e44f3af1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ed/76/104b2aa106a208da8b17a2fb72e033a5a9d7073c68f7e508b94916ed47a9/markupsafe-3.0.3-cp313-cp313-musllinux_1_2_riscv64.whl", hash = "sha256:795e7751525cae078558e679d646ae45574b47ed6e7771863fcc079a6171a0fc" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b5/99/16a5eb2d140087ebd97180d95249b00a03aa87e29cc224056274f2e45fd6/markupsafe-3.0.3-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:8485f406a96febb5140bfeca44a73e3ce5116b2501ac54fe953e488fb1d03b12" },
{ url = "https://mirrors.aliyun.com/pypi/packages/19/bc/e7140ed90c5d61d77cea142eed9f9c303f4c4806f60a1044c13e3f1471d0/markupsafe-3.0.3-cp313-cp313-win32.whl", hash = "sha256:bdd37121970bfd8be76c5fb069c7751683bdf373db1ed6c010162b2a130248ed" },
{ url = "https://mirrors.aliyun.com/pypi/packages/05/73/c4abe620b841b6b791f2edc248f556900667a5a1cf023a6646967ae98335/markupsafe-3.0.3-cp313-cp313-win_amd64.whl", hash = "sha256:9a1abfdc021a164803f4d485104931fb8f8c1efd55bc6b748d2f5774e78b62c5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f0/3a/fa34a0f7cfef23cf9500d68cb7c32dd64ffd58a12b09225fb03dd37d5b80/markupsafe-3.0.3-cp313-cp313-win_arm64.whl", hash = "sha256:7e68f88e5b8799aa49c85cd116c932a1ac15caaa3f5db09087854d218359e485" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e4/d7/e05cd7efe43a88a17a37b3ae96e79a19e846f3f456fe79c57ca61356ef01/markupsafe-3.0.3-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:218551f6df4868a8d527e3062d0fb968682fe92054e89978594c28e642c43a73" },
{ url = "https://mirrors.aliyun.com/pypi/packages/99/9e/e412117548182ce2148bdeacdda3bb494260c0b0184360fe0d56389b523b/markupsafe-3.0.3-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:3524b778fe5cfb3452a09d31e7b5adefeea8c5be1d43c4f810ba09f2ceb29d37" },
{ url = "https://mirrors.aliyun.com/pypi/packages/bc/e6/fa0ffcda717ef64a5108eaa7b4f5ed28d56122c9a6d70ab8b72f9f715c80/markupsafe-3.0.3-cp313-cp313t-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:4e885a3d1efa2eadc93c894a21770e4bc67899e3543680313b09f139e149ab19" },
{ url = "https://mirrors.aliyun.com/pypi/packages/96/ec/2102e881fe9d25fc16cb4b25d5f5cde50970967ffa5dddafdb771237062d/markupsafe-3.0.3-cp313-cp313t-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:8709b08f4a89aa7586de0aadc8da56180242ee0ada3999749b183aa23df95025" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4b/30/6f2fce1f1f205fc9323255b216ca8a235b15860c34b6798f810f05828e32/markupsafe-3.0.3-cp313-cp313t-manylinux_2_31_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:b8512a91625c9b3da6f127803b166b629725e68af71f8184ae7e7d54686a56d6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/58/47/4a0ccea4ab9f5dcb6f79c0236d954acb382202721e704223a8aafa38b5c8/markupsafe-3.0.3-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:9b79b7a16f7fedff2495d684f2b59b0457c3b493778c9eed31111be64d58279f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/6a/70/3780e9b72180b6fecb83a4814d84c3bf4b4ae4bf0b19c27196104149734c/markupsafe-3.0.3-cp313-cp313t-musllinux_1_2_riscv64.whl", hash = "sha256:12c63dfb4a98206f045aa9563db46507995f7ef6d83b2f68eda65c307c6829eb" },
{ url = "https://mirrors.aliyun.com/pypi/packages/98/c5/c03c7f4125180fc215220c035beac6b9cb684bc7a067c84fc69414d315f5/markupsafe-3.0.3-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:8f71bc33915be5186016f675cd83a1e08523649b0e33efdb898db577ef5bb009" },
{ url = "https://mirrors.aliyun.com/pypi/packages/80/d6/2d1b89f6ca4bff1036499b1e29a1d02d282259f3681540e16563f27ebc23/markupsafe-3.0.3-cp313-cp313t-win32.whl", hash = "sha256:69c0b73548bc525c8cb9a251cddf1931d1db4d2258e9599c28c07ef3580ef354" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2b/98/e48a4bfba0a0ffcf9925fe2d69240bfaa19c6f7507b8cd09c70684a53c1e/markupsafe-3.0.3-cp313-cp313t-win_amd64.whl", hash = "sha256:1b4b79e8ebf6b55351f0d91fe80f893b4743f104bff22e90697db1590e47a218" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0e/72/e3cc540f351f316e9ed0f092757459afbc595824ca724cbc5a5d4263713f/markupsafe-3.0.3-cp313-cp313t-win_arm64.whl", hash = "sha256:ad2cf8aa28b8c020ab2fc8287b0f823d0a7d8630784c31e9ee5edea20f406287" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "matplotlib"
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
version = "3.10.8"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
Bump infinity to v0.6.11. Requires python>=3.11 (#11814) ### What problem does this PR solve? Bump infinity to v0.6.11. Requires python>=3.11 ### Type of change - [x] Refactoring
2025-12-09 16:23:37 +08:00
{ name = "contourpy" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "cycler" },
{ name = "fonttools" },
{ name = "kiwisolver" },
{ name = "numpy" },
{ name = "packaging" },
{ name = "pillow" },
{ name = "pyparsing" },
{ name = "python-dateutil" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/8a/76/d3c6e3a13fe484ebe7718d14e269c9569c4eb0020a968a327acb3b9a8fe6/matplotlib-3.10.8.tar.gz", hash = "sha256:2299372c19d56bcd35cf05a2738308758d32b9eaed2371898d8f5bd33f084aa3" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/3d/b9/15fd5541ef4f5b9a17eefd379356cf12175fe577424e7b1d80676516031a/matplotlib-3.10.8-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:3f2e409836d7f5ac2f1c013110a4d50b9f7edc26328c108915f9075d7d7a91b6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8d/a0/2ba3473c1b66b9c74dc7107c67e9008cb1782edbe896d4c899d39ae9cf78/matplotlib-3.10.8-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:56271f3dac49a88d7fca5060f004d9d22b865f743a12a23b1e937a0be4818ee1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/75/97/a471f1c3eb1fd6f6c24a31a5858f443891d5127e63a7788678d14e249aea/matplotlib-3.10.8-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:a0a7f52498f72f13d4a25ea70f35f4cb60642b466cbb0a9be951b5bc3f45a486" },
{ url = "https://mirrors.aliyun.com/pypi/packages/01/be/cd478f4b66f48256f42927d0acbcd63a26a893136456cd079c0cc24fbabf/matplotlib-3.10.8-cp313-cp313-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:646d95230efb9ca614a7a594d4fcacde0ac61d25e37dd51710b36477594963ce" },
{ url = "https://mirrors.aliyun.com/pypi/packages/5d/7c/8dc289776eae5109e268c4fb92baf870678dc048a25d4ac903683b86d5bf/matplotlib-3.10.8-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:f89c151aab2e2e23cb3fe0acad1e8b82841fd265379c4cecd0f3fcb34c15e0f6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/64/40/37612487cc8a437d4dd261b32ca21fe2d79510fe74af74e1f42becb1bdb8/matplotlib-3.10.8-cp313-cp313-win_amd64.whl", hash = "sha256:e8ea3e2d4066083e264e75c829078f9e149fa119d27e19acd503de65e0b13149" },
{ url = "https://mirrors.aliyun.com/pypi/packages/66/52/8d8a8730e968185514680c2a6625943f70269509c3dcfc0dcf7d75928cb8/matplotlib-3.10.8-cp313-cp313-win_arm64.whl", hash = "sha256:c108a1d6fa78a50646029cb6d49808ff0fc1330fda87fa6f6250c6b5369b6645" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b5/27/51fe26e1062f298af5ef66343d8ef460e090a27fea73036c76c35821df04/matplotlib-3.10.8-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:ad3d9833a64cf48cc4300f2b406c3d0f4f4724a91c0bd5640678a6ba7c102077" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2c/1e/4de865bc591ac8e3062e835f42dd7fe7a93168d519557837f0e37513f629/matplotlib-3.10.8-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:eb3823f11823deade26ce3b9f40dcb4a213da7a670013929f31d5f5ed1055b22" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c6/cb/2f7b6e75fb4dce87ef91f60cac4f6e34f4c145ab036a22318ec837971300/matplotlib-3.10.8-cp313-cp313t-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:d9050fee89a89ed57b4fb2c1bfac9a3d0c57a0d55aed95949eedbc42070fea39" },
{ url = "https://mirrors.aliyun.com/pypi/packages/46/b3/bd9c57d6ba670a37ab31fb87ec3e8691b947134b201f881665b28cc039ff/matplotlib-3.10.8-cp313-cp313t-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:b44d07310e404ba95f8c25aa5536f154c0a8ec473303535949e52eb71d0a1565" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c0/3d/8b94a481456dfc9dfe6e39e93b5ab376e50998cddfd23f4ae3b431708f16/matplotlib-3.10.8-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:0a33deb84c15ede243aead39f77e990469fff93ad1521163305095b77b72ce4a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/bd/cd/bc06149fe5585ba800b189a6a654a75f1f127e8aab02fd2be10df7fa500c/matplotlib-3.10.8-cp313-cp313t-win_amd64.whl", hash = "sha256:3a48a78d2786784cc2413e57397981fb45c79e968d99656706018d6e62e57958" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e3/de/b22cf255abec916562cc04eef457c13e58a1990048de0c0c3604d082355e/matplotlib-3.10.8-cp313-cp313t-win_arm64.whl", hash = "sha256:15d30132718972c2c074cd14638c7f4592bd98719e2308bccea40e0538bc0cb5" },
Bump python to >=3.12 (#11846) ### What problem does this PR solve? Bump python to >=3.12 ### Type of change - [x] Refactoring
2025-12-09 19:55:25 +08:00
]
Feat: mcp server (#7084) ### What problem does this PR solve? Add MCP support with a client example. Issue link: #4344 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-04-21 09:43:20 +08:00
[[package]]
name = "mcp"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "1.19.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: mcp server (#7084) ### What problem does this PR solve? Add MCP support with a client example. Issue link: #4344 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-04-21 09:43:20 +08:00
dependencies = [
{ name = "anyio" },
{ name = "httpx" },
{ name = "httpx-sse" },
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
{ name = "jsonschema" },
Feat: mcp server (#7084) ### What problem does this PR solve? Add MCP support with a client example. Issue link: #4344 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-04-21 09:43:20 +08:00
{ name = "pydantic" },
{ name = "pydantic-settings" },
Refa: upgrade MCP SDK to v1.9.4 (#8421) ### What problem does this PR solve? Upgrade MCP SDK to v1.9.4 (latest). ### Type of change - [x] Refactoring
2025-06-23 16:53:59 +08:00
{ name = "python-multipart" },
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
{ name = "pywin32", marker = "sys_platform == 'win32'" },
Feat: mcp server (#7084) ### What problem does this PR solve? Add MCP support with a client example. Issue link: #4344 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-04-21 09:43:20 +08:00
{ name = "sse-starlette" },
{ name = "starlette" },
Refa: upgrade MCP SDK to v1.9.4 (#8421) ### What problem does this PR solve? Upgrade MCP SDK to v1.9.4 (latest). ### Type of change - [x] Refactoring
2025-06-23 16:53:59 +08:00
{ name = "uvicorn", marker = "sys_platform != 'emscripten'" },
Feat: mcp server (#7084) ### What problem does this PR solve? Add MCP support with a client example. Issue link: #4344 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-04-21 09:43:20 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/69/2b/916852a5668f45d8787378461eaa1244876d77575ffef024483c94c0649c/mcp-1.19.0.tar.gz", hash = "sha256:213de0d3cd63f71bc08ffe9cc8d4409cc87acffd383f6195d2ce0457c021b5c1" }
Feat: mcp server (#7084) ### What problem does this PR solve? Add MCP support with a client example. Issue link: #4344 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-04-21 09:43:20 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ce/a3/3e71a875a08b6a830b88c40bc413bff01f1650f1efe8a054b5e90a9d4f56/mcp-1.19.0-py3-none-any.whl", hash = "sha256:f5907fe1c0167255f916718f376d05f09a830a215327a3ccdd5ec8a519f2e572" },
Feat: mcp server (#7084) ### What problem does this PR solve? Add MCP support with a client example. Issue link: #4344 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-04-21 09:43:20 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "mdurl"
version = "0.1.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/d6/54/cfe61301667036ec958cb99bd3efefba235e65cdeb9c84d24a8293ba1d90/mdurl-0.1.2.tar.gz", hash = "sha256:bb413d29f5eea38f31dd4754dd7377d4465116fb207585f97bf925588687c1ba" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/b3/38/89ba8ad64ae25be8de66a6d463314cf1eb366222074cfda9ee839c56a4b4/mdurl-0.1.2-py3-none-any.whl", hash = "sha256:84008a41e51615a49fc9966191ff91509e3c40b939176e643fd50a5c2196b8f8" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "mini-racer"
version = "0.12.4"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/8c/2d/e051f58e17117b1b8b11a7d17622c1528fa9002c553943c6b677c1b412da/mini_racer-0.12.4.tar.gz", hash = "sha256:84c67553ce9f3736d4c617d8a3f882949d37a46cfb47fe11dab33dd6704e62a4" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/71/fe/1452b6c74cae9e8cd7b6a16d8b1ef08bba4dd0ed373a95f3b401c2e712ea/mini_racer-0.12.4-py3-none-macosx_10_9_x86_64.whl", hash = "sha256:bce8a3cee946575a352f5e65335903bc148da42c036d0c738ac67e931600e455" },
{ url = "https://mirrors.aliyun.com/pypi/packages/99/ae/c22478eff26e6136341e6b40d34f8d285f910ca4d2e2a0ca4703ef87be79/mini_racer-0.12.4-py3-none-macosx_11_0_arm64.whl", hash = "sha256:56c832e6ac2db6a304d1e8e80030615297aafbc6940f64f3479af4ba16abccd5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/44/89/f062aa116b14fcace91f0af86a37605f0ba7c07a01c8101b5ea104d489b1/mini_racer-0.12.4-py3-none-manylinux_2_31_aarch64.whl", hash = "sha256:b82c4bd2976e280ed0a72c9c2de01b13f18ccfbe6f4892cbc22aae04410fac3c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9c/a1/09122c88a0dd0a2141b0ea068d70f5d31acd0015d6f3157b8efd3ff7e026/mini_racer-0.12.4-py3-none-manylinux_2_31_x86_64.whl", hash = "sha256:69a1c44d02a9069b881684cef15a2d747fe0743df29eadc881fda7002aae5fd2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/6c/3b/826e41f92631560e5c6ca2aa4ef9005bdccf9290c1e7ddebe05e0a3b8c7c/mini_racer-0.12.4-py3-none-musllinux_1_2_aarch64.whl", hash = "sha256:499dbc267dfe60e954bc1b6c3787f7b10fc41fe1975853c9a6ddb55eb83dc4d9" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e5/37/15b30316630d1f63b025f058dc92efa75931a37315c34ca07f80be2cc405/mini_racer-0.12.4-py3-none-musllinux_1_2_x86_64.whl", hash = "sha256:231f949f5787d18351939f1fe59e5a6fe134bccb5ecf8f836b9beab69d91c8d9" },
{ url = "https://mirrors.aliyun.com/pypi/packages/5c/0e/a9943f90b4a8a6d3849b81a00a00d2db128d876365385af382a0e2caf191/mini_racer-0.12.4-py3-none-win_amd64.whl", hash = "sha256:9446e3bd6a4eb9fbedf1861326f7476080995a31c9b69308acef17e5b7ecaa1b" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "minio"
version = "7.2.4"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "argon2-cffi" },
{ name = "certifi" },
{ name = "pycryptodome" },
{ name = "typing-extensions" },
{ name = "urllib3" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/3a/2e/7bd24eb2e02a19a03bd0e73e59c051c62c62cabdd305ccbc59a90143752c/minio-7.2.4.tar.gz", hash = "sha256:d504d8464e5198fb74dd9b572cc88b185ae7997c17705e8c09f3fef2f439d984" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/a8/29/17ec9cecedad692cf18abd0b5e57d7008d1dda8929915e7cfee76ea0e849/minio-7.2.4-py3-none-any.whl", hash = "sha256:91b51c21d25e3ee6d51f52eab126d6c974371add0d77951e42c322a59c5533e7" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "mistralai"
version = "0.4.2"
Fix: switch Python package mirror from Tsinghua to Aliyun (#13617) ### What problem does this PR solve? Replace pypi.tuna.tsinghua.edu.cn with mirrors.aliyun.com to resolve issues with missing packages on the Tsinghua mirror. ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2026-03-16 12:12:25 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "httpx" },
{ name = "orjson" },
{ name = "pydantic" },
]
Security: do not use litellm 1.82.7 and 1.82.8 (#13768) ### What problem does this PR solve? See [issue](https://github.com/BerriAI/litellm/issues/24518) from Litellm. Upgraded from `1.81.15` to `1.82.6`, so RAGFlow is safe as always. ### Type of change - [x] Security Co-authored-by: Jin Hai <haijin.chn@gmail.com> Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
2026-03-25 22:39:33 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/fa/20/4204f461588310b3a7ffbbbb7fa573493dc1c8185d376ee72516c04575bf/mistralai-0.4.2.tar.gz", hash = "sha256:5eb656710517168ae053f9847b0bb7f617eda07f1f93f946ad6c91a4d407fd93" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
Security: do not use litellm 1.82.7 and 1.82.8 (#13768) ### What problem does this PR solve? See [issue](https://github.com/BerriAI/litellm/issues/24518) from Litellm. Upgraded from `1.81.15` to `1.82.6`, so RAGFlow is safe as always. ### Type of change - [x] Security Co-authored-by: Jin Hai <haijin.chn@gmail.com> Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
2026-03-25 22:39:33 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/4f/fe/79dad76b8d94b62d9e2aab8446183190e1dc384c617d06c3c93307850e11/mistralai-0.4.2-py3-none-any.whl", hash = "sha256:63c98eea139585f0a3b2c4c6c09c453738bac3958055e6f2362d3866e96b0168" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "mistune"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "3.2.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/9d/55/d01f0c4b45ade6536c51170b9043db8b2ec6ddf4a35c7ea3f5f559ac935b/mistune-3.2.0.tar.gz", hash = "sha256:708487c8a8cdd99c9d90eb3ed4c3ed961246ff78ac82f03418f5183ab70e398a" }
Add Moodle data source integration (#11325) ### What problem does this PR solve? This PR adds a native Moodle connector to sync content (courses, resources, forums, assignments, pages, books) into RAGFlow. ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2025-11-21 12:58:49 +01:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/9b/f7/4a5e785ec9fbd65146a27b6b70b6cdc161a66f2024e4b04ac06a67f5578b/mistune-3.2.0-py3-none-any.whl", hash = "sha256:febdc629a3c78616b94393c6580551e0e34cc289987ec6c35ed3f4be42d0eee1" },
Add Moodle data source integration (#11325) ### What problem does this PR solve? This PR adds a native Moodle connector to sync content (courses, resources, forums, assignments, pages, books) into RAGFlow. ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2025-11-21 12:58:49 +01:00
]
[[package]]
name = "moodlepy"
version = "0.24.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Add Moodle data source integration (#11325) ### What problem does this PR solve? This PR adds a native Moodle connector to sync content (courses, resources, forums, assignments, pages, books) into RAGFlow. ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2025-11-21 12:58:49 +01:00
dependencies = [
{ name = "attrs" },
{ name = "cattrs" },
{ name = "requests" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/69/83/d03072735fc822b225efefa7f7646a4ed6cd73d1c717a338871b9958ce5d/moodlepy-0.24.1.tar.gz", hash = "sha256:94d361e4da56748d29910e01979e4652a42220994112b4f07589f200cb7915e3" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/18/6b/c526e3230e20171d7791f0fb2137aa4b49c82d9878959f089be1162a7c72/moodlepy-0.24.1-py3-none-any.whl", hash = "sha256:2809ece7a167d7ecc2a744cde188f0af66e1db58863e7a2ed77d1a0b08ff82e2" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "mpmath"
version = "1.3.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/e0/47/dd32fa426cc72114383ac549964eecb20ecfd886d1e5ccf5340b55b02f57/mpmath-1.3.0.tar.gz", hash = "sha256:7a28eb2a9774d00c7bc92411c19a89209d5da7c4c9a9e227be8330a23a25b91f" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/43/e3/7d92a15f894aa0c9c4b49b8ee9ac9850d6e63b03c9c32c0367a13ae62209/mpmath-1.3.0-py3-none-any.whl", hash = "sha256:a0b2b9fe80bbcd81a6647ff13108738cfb482d481d826cc0e02f5b35e5c88d2c" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "msal"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.37.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "cryptography" },
{ name = "pyjwt", extra = ["crypto"] },
{ name = "requests" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/9a/99/d840198ecf6e8057bbc937f129ae940404485d736cda73253bbff9537f01/msal-1.37.0.tar.gz", hash = "sha256:1b1672a33ee467c1d70b341bb16cafd51bb3c817147a95b93263794b03971bec" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/94/b0/d807279f4b55d16d1f120d5ac4344c6e39b56732e2a224d40bded7fd67ad/msal-1.37.0-py3-none-any.whl", hash = "sha256:dd17e95a7c71bce75e8108113438ba7c4a086b3bcad4f57a8c09b7af3d753c2d" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "msal-extensions"
Feat: add basic Langfuse support for LLM module (#6443) ### What problem does this PR solve? #6155 Add basic Langfuse support for LLM module. A trace example: <img width="755" alt="image" src="https://github.com/user-attachments/assets/25c1f852-5116-486c-a47f-6097187142ca" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-03-24 13:18:47 +08:00
version = "1.3.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "msal" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/01/99/5d239b6156eddf761a636bded1118414d161bd6b7b37a9335549ed159396/msal_extensions-1.3.1.tar.gz", hash = "sha256:c5b0fd10f65ef62b5f1d62f4251d51cbcaf003fcedae8c91b040a488614be1a4" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/5e/75/bd9b7bb966668920f06b200e84454c8f3566b102183bc55c5473d96cb2b9/msal_extensions-1.3.1-py3-none-any.whl", hash = "sha256:96d3de4d034504e969ac5e85bae8106c8373b5c6568e4c8fa7af2eca9dbe6bca" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "msgspec"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "0.20.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ea/9c/bfbd12955a49180cbd234c5d29ec6f74fe641698f0cd9df154a854fc8a15/msgspec-0.20.0.tar.gz", hash = "sha256:692349e588fde322875f8d3025ac01689fead5901e7fb18d6870a44519d62a29" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/8a/d1/b902d38b6e5ba3bdddbec469bba388d647f960aeed7b5b3623a8debe8a76/msgspec-0.20.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:9c1ff8db03be7598b50dd4b4a478d6fe93faae3bd54f4f17aa004d0e46c14c46" },
{ url = "https://mirrors.aliyun.com/pypi/packages/57/b6/eff0305961a1d9447ec2b02f8c73c8946f22564d302a504185b730c9a761/msgspec-0.20.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:f6532369ece217fd37c5ebcfd7e981f2615628c21121b7b2df9d3adcf2fd69b8" },
{ url = "https://mirrors.aliyun.com/pypi/packages/99/93/f2ec1ae1de51d3fdee998a1ede6b2c089453a2ee82b5c1b361ed9095064a/msgspec-0.20.0-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:f9a1697da2f85a751ac3cc6a97fceb8e937fc670947183fb2268edaf4016d1ee" },
{ url = "https://mirrors.aliyun.com/pypi/packages/28/83/36557b04cfdc317ed8a525c4993b23e43a8fbcddaddd78619112ca07138c/msgspec-0.20.0-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:7fac7e9c92eddcd24c19d9e5f6249760941485dff97802461ae7c995a2450111" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8f/56/362037a1ed5be0b88aced59272442c4b40065c659700f4b195a7f4d0ac88/msgspec-0.20.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:f953a66f2a3eb8d5ea64768445e2bb301d97609db052628c3e1bcb7d87192a9f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/92/75/fa2370ec341cedf663731ab7042e177b3742645c5dd4f64dc96bd9f18a6b/msgspec-0.20.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:247af0313ae64a066d3aea7ba98840f6681ccbf5c90ba9c7d17f3e39dbba679c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f1/25/5e8080fe0117f799b1b68008dc29a65862077296b92550632de015128579/msgspec-0.20.0-cp313-cp313-win_amd64.whl", hash = "sha256:67d5e4dfad52832017018d30a462604c80561aa62a9d548fc2bd4e430b66a352" },
{ url = "https://mirrors.aliyun.com/pypi/packages/79/b6/63363422153937d40e1cb349c5081338401f8529a5a4e216865decd981bf/msgspec-0.20.0-cp313-cp313-win_arm64.whl", hash = "sha256:91a52578226708b63a9a13de287b1ec3ed1123e4a088b198143860c087770458" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
[[package]]
name = "msoffcrypto-tool"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "6.0.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
dependencies = [
{ name = "cryptography" },
{ name = "olefile" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a6/34/6250bdddaeaae24098e45449ea362fb3555a65fba30cad0ad5630ea48d1a/msoffcrypto_tool-6.0.0.tar.gz", hash = "sha256:9a5ebc4c0096b42e5d7ebc2350afdc92dc511061e935ca188468094fdd032bbe" }
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/3c/85/9e359fa9279e1d6861faaf9b6f037a3226374deb20a054c3937be6992013/msoffcrypto_tool-6.0.0-py3-none-any.whl", hash = "sha256:46c394ed5d9641e802fc79bf3fb0666a53748b23fa8c4aa634ae9d30d46fe397" },
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "multidict"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "6.7.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/80/1e/5492c365f222f907de1039b91f922b93fa4f764c713ee858d235495d8f50/multidict-6.7.0.tar.gz", hash = "sha256:c6e99d9a65ca282e578dfea819cfa9c0a62b2499d8677392e09feaf305e9e6f5" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/d2/86/33272a544eeb36d66e4d9a920602d1a2f57d4ebea4ef3cdfe5a912574c95/multidict-6.7.0-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:bee7c0588aa0076ce77c0ea5d19a68d76ad81fcd9fe8501003b9a24f9d4000f6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/91/1c/eb97db117a1ebe46d457a3d235a7b9d2e6dcab174f42d1b67663dd9e5371/multidict-6.7.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:7ef6b61cad77091056ce0e7ce69814ef72afacb150b7ac6a3e9470def2198159" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f1/d8/6c3442322e41fb1dd4de8bd67bfd11cd72352ac131f6368315617de752f1/multidict-6.7.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:9c0359b1ec12b1d6849c59f9d319610b7f20ef990a6d454ab151aa0e3b9f78ca" },
{ url = "https://mirrors.aliyun.com/pypi/packages/75/3f/e2639e80325af0b6c6febdf8e57cc07043ff15f57fa1ef808f4ccb5ac4cd/multidict-6.7.0-cp313-cp313-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:cd240939f71c64bd658f186330603aac1a9a81bf6273f523fca63673cb7378a8" },
{ url = "https://mirrors.aliyun.com/pypi/packages/5d/cc/84e0585f805cbeaa9cbdaa95f9a3d6aed745b9d25700623ac89a6ecff400/multidict-6.7.0-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:a60a4d75718a5efa473ebd5ab685786ba0c67b8381f781d1be14da49f1a2dc60" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b0/9c/ac851c107c92289acbbf5cfb485694084690c1b17e555f44952c26ddc5bd/multidict-6.7.0-cp313-cp313-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:53a42d364f323275126aff81fb67c5ca1b7a04fda0546245730a55c8c5f24bc4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/50/cc/5f93e99427248c09da95b62d64b25748a5f5c98c7c2ab09825a1d6af0e15/multidict-6.7.0-cp313-cp313-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:3b29b980d0ddbecb736735ee5bef69bb2ddca56eff603c86f3f29a1128299b4f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ec/0c/2ec1d883ceb79c6f7f6d7ad90c919c898f5d1c6ea96d322751420211e072/multidict-6.7.0-cp313-cp313-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:f8a93b1c0ed2d04b97a5e9336fd2d33371b9a6e29ab7dd6503d63407c20ffbaf" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c6/2d/f0b184fa88d6630aa267680bdb8623fb69cb0d024b8c6f0d23f9a0f406d3/multidict-6.7.0-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:9ff96e8815eecacc6645da76c413eb3b3d34cfca256c70b16b286a687d013c32" },
{ url = "https://mirrors.aliyun.com/pypi/packages/06/c9/11ea263ad0df7dfabcad404feb3c0dd40b131bc7f232d5537f2fb1356951/multidict-6.7.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:7516c579652f6a6be0e266aec0acd0db80829ca305c3d771ed898538804c2036" },
{ url = "https://mirrors.aliyun.com/pypi/packages/41/88/d714b86ee2c17d6e09850c70c9d310abac3d808ab49dfa16b43aba9d53fd/multidict-6.7.0-cp313-cp313-musllinux_1_2_armv7l.whl", hash = "sha256:040f393368e63fb0f3330e70c26bfd336656bed925e5cbe17c9da839a6ab13ec" },
{ url = "https://mirrors.aliyun.com/pypi/packages/15/fe/ad407bb9e818c2b31383f6131ca19ea7e35ce93cf1310fce69f12e89de75/multidict-6.7.0-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:b3bc26a951007b1057a1c543af845f1c7e3e71cc240ed1ace7bf4484aa99196e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8c/a4/a89abdb0229e533fb925e7c6e5c40201c2873efebc9abaf14046a4536ee6/multidict-6.7.0-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:7b022717c748dd1992a83e219587aabe45980d88969f01b316e78683e6285f64" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8d/aa/0e2b27bd88b40a4fb8dc53dd74eecac70edaa4c1dd0707eb2164da3675b3/multidict-6.7.0-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:9600082733859f00d79dee64effc7aef1beb26adb297416a4ad2116fd61374bd" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d0/8e/0c67b7120d5d5f6d874ed85a085f9dc770a7f9d8813e80f44a9fec820bb7/multidict-6.7.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:94218fcec4d72bc61df51c198d098ce2b378e0ccbac41ddbed5ef44092913288" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ba/55/b73e1d624ea4b8fd4dd07a3bb70f6e4c7c6c5d9d640a41c6ffe5cdbd2a55/multidict-6.7.0-cp313-cp313-win32.whl", hash = "sha256:a37bd74c3fa9d00be2d7b8eca074dc56bd8077ddd2917a839bd989612671ed17" },
{ url = "https://mirrors.aliyun.com/pypi/packages/32/31/75c59e7d3b4205075b4c183fa4ca398a2daf2303ddf616b04ae6ef55cffe/multidict-6.7.0-cp313-cp313-win_amd64.whl", hash = "sha256:30d193c6cc6d559db42b6bcec8a5d395d34d60c9877a0b71ecd7c204fcf15390" },
{ url = "https://mirrors.aliyun.com/pypi/packages/31/2a/8987831e811f1184c22bc2e45844934385363ee61c0a2dcfa8f71b87e608/multidict-6.7.0-cp313-cp313-win_arm64.whl", hash = "sha256:ea3334cabe4d41b7ccd01e4d349828678794edbc2d3ae97fc162a3312095092e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e8/68/7b3a5170a382a340147337b300b9eb25a9ddb573bcdfff19c0fa3f31ffba/multidict-6.7.0-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:ad9ce259f50abd98a1ca0aa6e490b58c316a0fce0617f609723e40804add2c00" },
{ url = "https://mirrors.aliyun.com/pypi/packages/55/5c/3fa2d07c84df4e302060f555bbf539310980362236ad49f50eeb0a1c1eb9/multidict-6.7.0-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:07f5594ac6d084cbb5de2df218d78baf55ef150b91f0ff8a21cc7a2e3a5a58eb" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fc/56/67212d33239797f9bd91962bb899d72bb0f4c35a8652dcdb8ed049bef878/multidict-6.7.0-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:0591b48acf279821a579282444814a2d8d0af624ae0bc600aa4d1b920b6e924b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/46/d1/908f896224290350721597a61a69cd19b89ad8ee0ae1f38b3f5cd12ea2ac/multidict-6.7.0-cp313-cp313t-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:749a72584761531d2b9467cfbdfd29487ee21124c304c4b6cb760d8777b27f9c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ab/67/8604288bbd68680eee0ab568fdcb56171d8b23a01bcd5cb0c8fedf6e5d99/multidict-6.7.0-cp313-cp313t-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:6b4c3d199f953acd5b446bf7c0de1fe25d94e09e79086f8dc2f48a11a129cdf1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/20/33/9228d76339f1ba51e3efef7da3ebd91964d3006217aae13211653193c3ff/multidict-6.7.0-cp313-cp313t-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:9fb0211dfc3b51efea2f349ec92c114d7754dd62c01f81c3e32b765b70c45c9b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f8/2d/25d9b566d10cab1c42b3b9e5b11ef79c9111eaf4463b8c257a3bd89e0ead/multidict-6.7.0-cp313-cp313t-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:a027ec240fe73a8d6281872690b988eed307cd7d91b23998ff35ff577ca688b5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b6/b1/8d1a965e6637fc33de3c0d8f414485c2b7e4af00f42cab3d84e7b955c222/multidict-6.7.0-cp313-cp313t-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:d1d964afecdf3a8288789df2f5751dc0a8261138c3768d9af117ed384e538fad" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ba/0c/06b5a8adbdeedada6f4fb8d8f193d44a347223b11939b42953eeb6530b6b/multidict-6.7.0-cp313-cp313t-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:caf53b15b1b7df9fbd0709aa01409000a2b4dd03a5f6f5cc548183c7c8f8b63c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8f/31/b2491b5fe167ca044c6eb4b8f2c9f3b8a00b24c432c365358eadac5d7625/multidict-6.7.0-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:654030da3197d927f05a536a66186070e98765aa5142794c9904555d3a9d8fb5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/61/1a/982913957cb90406c8c94f53001abd9eafc271cb3e70ff6371590bec478e/multidict-6.7.0-cp313-cp313t-musllinux_1_2_armv7l.whl", hash = "sha256:2090d3718829d1e484706a2f525e50c892237b2bf9b17a79b059cb98cddc2f10" },
{ url = "https://mirrors.aliyun.com/pypi/packages/be/c0/21435d804c1a1cf7a2608593f4d19bca5bcbd7a81a70b253fdd1c12af9c0/multidict-6.7.0-cp313-cp313t-musllinux_1_2_i686.whl", hash = "sha256:2d2cfeec3f6f45651b3d408c4acec0ebf3daa9bc8a112a084206f5db5d05b754" },
{ url = "https://mirrors.aliyun.com/pypi/packages/54/0a/4349d540d4a883863191be6eb9a928846d4ec0ea007d3dcd36323bb058ac/multidict-6.7.0-cp313-cp313t-musllinux_1_2_ppc64le.whl", hash = "sha256:4ef089f985b8c194d341eb2c24ae6e7408c9a0e2e5658699c92f497437d88c3c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/26/64/d5416038dbda1488daf16b676e4dbfd9674dde10a0cc8f4fc2b502d8125d/multidict-6.7.0-cp313-cp313t-musllinux_1_2_s390x.whl", hash = "sha256:e93a0617cd16998784bf4414c7e40f17a35d2350e5c6f0bd900d3a8e02bd3762" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9f/8c/8290c50d14e49f35e0bd4abc25e1bc7711149ca9588ab7d04f886cdf03d9/multidict-6.7.0-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:f0feece2ef8ebc42ed9e2e8c78fc4aa3cf455733b507c09ef7406364c94376c6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ef/a0/f83ae75e42d694b3fbad3e047670e511c138be747bc713cf1b10d5096416/multidict-6.7.0-cp313-cp313t-win32.whl", hash = "sha256:19a1d55338ec1be74ef62440ca9e04a2f001a04d0cc49a4983dc320ff0f3212d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/dc/80/9b174a92814a3830b7357307a792300f42c9e94664b01dee8e457551fa66/multidict-6.7.0-cp313-cp313t-win_amd64.whl", hash = "sha256:3da4fb467498df97e986af166b12d01f05d2e04f978a9c1c680ea1988e0bc4b6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/cc/28/04baeaf0428d95bb7a7bea0e691ba2f31394338ba424fb0679a9ed0f4c09/multidict-6.7.0-cp313-cp313t-win_arm64.whl", hash = "sha256:b4121773c49a0776461f4a904cdf6264c88e42218aaa8407e803ca8025872792" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b7/da/7d22601b625e241d4f23ef1ebff8acfc60da633c9e7e7922e24d10f592b3/multidict-6.7.0-py3-none-any.whl", hash = "sha256:394fc5c42a333c9ffc3e421a4c85e08580d990e08b99f6bf35b4132114c5dcb3" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "multiprocess"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "0.70.18"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "dill" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/72/fd/2ae3826f5be24c6ed87266bc4e59c46ea5b059a103f3d7e7eb76a52aeecb/multiprocess-0.70.18.tar.gz", hash = "sha256:f9597128e6b3e67b23956da07cf3d2e5cba79e2f4e0fba8d7903636663ec6d0d" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ba/d8/0cba6cf51a1a31f20471fbc823a716170c73012ddc4fb85d706630ed6e8f/multiprocess-0.70.18-py310-none-any.whl", hash = "sha256:60c194974c31784019c1f459d984e8f33ee48f10fcf42c309ba97b30d9bd53ea" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4b/88/9039f2fed1012ef584751d4ceff9ab4a51e5ae264898f0b7cbf44340a859/multiprocess-0.70.18-py311-none-any.whl", hash = "sha256:5aa6eef98e691281b3ad923be2832bf1c55dd2c859acd73e5ec53a66aae06a1d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/bf/b6/5f922792be93b82ec6b5f270bbb1ef031fd0622847070bbcf9da816502cc/multiprocess-0.70.18-py312-none-any.whl", hash = "sha256:9b78f8e5024b573730bfb654783a13800c2c0f2dfc0c25e70b40d184d64adaa2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ee/25/7d7e78e750bc1aecfaf0efbf826c69a791d2eeaf29cf20cba93ff4cced78/multiprocess-0.70.18-py313-none-any.whl", hash = "sha256:871743755f43ef57d7910a38433cfe41319e72be1bbd90b79c7a5ac523eb9334" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3b/c3/ca84c19bd14cdfc21c388fdcebf08b86a7a470ebc9f5c3c084fc2dbc50f7/multiprocess-0.70.18-py38-none-any.whl", hash = "sha256:dbf705e52a154fe5e90fb17b38f02556169557c2dd8bb084f2e06c2784d8279b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/6c/28/dd72947e59a6a8c856448a5e74da6201cb5502ddff644fbc790e4bd40b9a/multiprocess-0.70.18-py39-none-any.whl", hash = "sha256:e78ca805a72b1b810c690b6b4cc32579eba34f403094bbbae962b7b5bf9dfcb8" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "multitasking"
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
version = "0.0.12"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/17/0d/74f0293dfd7dcc3837746d0138cbedd60b31701ecc75caec7d3f281feba0/multitasking-0.0.12.tar.gz", hash = "sha256:2fba2fa8ed8c4b85e227c5dd7dc41c7d658de3b6f247927316175a57349b84d1" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
[[package]]
name = "murmurhash"
version = "1.0.15"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/23/2e/88c147931ea9725d634840d538622e94122bceaf346233349b7b5c62964b/murmurhash-1.0.15.tar.gz", hash = "sha256:58e2b27b7847f9e2a6edf10b47a8c8dd70a4705f45dccb7bf76aeadacf56ba01" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/29/2f/ba300b5f04dae0409202d6285668b8a9d3ade43a846abee3ef611cb388d5/murmurhash-1.0.15-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:fe50dc70e52786759358fd1471e309b94dddfffb9320d9dfea233c7684c894ba" },
{ url = "https://mirrors.aliyun.com/pypi/packages/34/02/29c19d268e6f4ea1ed2a462c901eed1ed35b454e2cbc57da592fad663ac6/murmurhash-1.0.15-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:1349a7c23f6092e7998ddc5bd28546cc31a595afc61e9fdb3afc423feec3d7ad" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e2/63/58e2de2b5232cd294c64092688c422196e74f9fa8b3958bdf02d33df24b9/murmurhash-1.0.15-cp313-cp313-manylinux1_x86_64.manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:b3ba6d05de2613535b5a9227d4ad8ef40a540465f64660d4a8800634ae10e04f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/aa/9a/d13e2e9f8ba1ced06840921a50f7cece0a475453284158a3018b72679761/murmurhash-1.0.15-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:fa1b70b3cc2801ab44179c65827bbd12009c68b34e9d9ce7125b6a0bd35af63c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b2/e1/47994f1813fa205c84977b0ff51ae6709f8539af052c7491a5f863d82bdc/murmurhash-1.0.15-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:213d710fb6f4ef3bc11abbfad0fa94a75ffb675b7dc158c123471e5de869f9af" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b9/ea/90c1fd00b4aeb704fb5e84cd666b33ffd7f245155048071ffbb51d2bb57d/murmurhash-1.0.15-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:b65a5c4e7f5d71f7ccac2d2b60bdf7092d7976270878cfec59d5a66a533db823" },
{ url = "https://mirrors.aliyun.com/pypi/packages/00/db/da73462dbfa77f6433b128d2120ba7ba300f8c06dc4f4e022c38d240a5f5/murmurhash-1.0.15-cp313-cp313-win_amd64.whl", hash = "sha256:9aba94c5d841e1904cd110e94ceb7f49cfb60a874bbfb27e0373622998fb7c7c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/bb/83/032729ef14971b938fbef41ee125fc8800020ee229bd35178b6ede8ee934/murmurhash-1.0.15-cp313-cp313-win_arm64.whl", hash = "sha256:263807eca40d08c7b702413e45cca75ecb5883aa337237dc5addb660f1483378" },
{ url = "https://mirrors.aliyun.com/pypi/packages/10/83/7547d9205e9bd2f8e5dfd0b682cc9277594f98909f228eb359489baec1df/murmurhash-1.0.15-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:694fd42a74b7ce257169d14c24aa616aa6cd4ccf8abe50eca0557e08da99d055" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b7/c7/3afd5de7a5b3ae07fe2d3a3271b327ee1489c58ba2b2f2159bd31a25edb9/murmurhash-1.0.15-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:a2ea4546ba426390beff3cd10db8f0152fdc9072c4f2583ec7d8aa9f3e4ac070" },
{ url = "https://mirrors.aliyun.com/pypi/packages/02/69/d6637ee67d78ebb2538c00411f28ea5c154886bbe1db16c49435a8a4ab16/murmurhash-1.0.15-cp313-cp313t-manylinux1_x86_64.manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:34e5a91139c40b10f98d0b297907f5d5267b4b1b2e5dd2eb74a021824f751b98" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ab/4c/89e590165b4c7da6bf941441212a721a270195332d3aacfdfdf527d466ca/murmurhash-1.0.15-cp313-cp313t-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:dc35606868a5961cf42e79314ca0bddf5a400ce377b14d83192057928d6252ec" },
{ url = "https://mirrors.aliyun.com/pypi/packages/07/7a/95c42df0c21d2e413b9fcd17317a7587351daeb264dc29c6aec1fdbd26f8/murmurhash-1.0.15-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:43cc6ac3b91ca0f7a5ae9c063ba4d6c26972c97fd7c25280ecc666413e4c5535" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d0/22/9d02c880a88b83bb3ce7d6a38fb727373ab78d82e5f3d8d9fc5612219f90/murmurhash-1.0.15-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:847d712136cb462f0e4bd6229ee2d9eb996d8854eb8312dff3d20c8f5181fda5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9a/e3/750232524e0dc262e8dcede6536dafc766faadd9a52f1d23746b02948ad8/murmurhash-1.0.15-cp313-cp313t-win_amd64.whl", hash = "sha256:2680851af6901dbe66cc4aa7ef8e263de47e6e1b425ae324caa571bdf18f8d58" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ff/89/4ad9d215ef6ade89f27a72dc4e86b98ef1a43534cc3e6a6900a362a0bf0a/murmurhash-1.0.15-cp313-cp313t-win_arm64.whl", hash = "sha256:189a8de4d657b5da9efd66601b0636330b08262b3a55431f2379097c986995d0" },
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "mygene"
version = "3.2.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "biothings-client" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/0a/ec/a256003f84196aa3fdd65a7c6f5adfc0688398fb66442eba75b39c9b7627/mygene-3.2.2.tar.gz", hash = "sha256:e729cabbc28cf5afb221bca1ab637883b375cb1a3e2f067587ec79f71affdaea" }
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/a7/b7/132b1673c0ec00881d49d56c09624942fa0ebd2fc21d73d80647efa082e9/mygene-3.2.2-py2.py3-none-any.whl", hash = "sha256:18d85d1b28ecee2be31d844607fb0c5f7d7c58573278432df819ee2a5e88fe46" },
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
]
[[package]]
name = "mypy-boto3-s3"
version = "1.40.26"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/00/b8/55d21ed9ca479df66d9892212ba7d7977850ef17aa80a83e3f11f31190fd/mypy_boto3_s3-1.40.26.tar.gz", hash = "sha256:8d2bfd1052894d0e84c9fb9358d838ba0eed0265076c7dd7f45622c770275c99" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/85/a5/dba3384423834009bdd41c7021de5c663468a0e7bc4071cb301721e52a99/mypy_boto3_s3-1.40.26-py3-none-any.whl", hash = "sha256:6d055d16ef89a0133ade92f6b4f09603e4acc31a0f5e8f846edf4eb48f17b5a7" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Refact: update description for max_token in embedding #12792 (#12845) ### What problem does this PR solve? Refact: update description for max_token in embedding #12792 ### Type of change - [x] Refactoring Co-authored-by: Liu An <asiro@qq.com>
2026-01-28 09:52:32 +08:00
[[package]]
name = "mysql-connector-python"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "9.3.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/82/5e/55b265cb95938e271208e5692d7e615c53f2aeea894ab72a9f14ab198e9a/mysql-connector-python-9.3.0.tar.gz", hash = "sha256:8b16d51447e3603f18478fb5a19b333bfb73fb58f872eb055a105635f53d2345" }
Fix: switch Python package mirror from Tsinghua to Aliyun (#13617) ### What problem does this PR solve? Replace pypi.tuna.tsinghua.edu.cn with mirrors.aliyun.com to resolve issues with missing packages on the Tsinghua mirror. ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2026-03-16 12:12:25 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/6a/16/5762061505a0d0d3a333613b6f5d7b8eb3222a689aa32f71ed15f1532ad1/mysql_connector_python-9.3.0-cp313-cp313-macosx_14_0_arm64.whl", hash = "sha256:9516a4cdbaee3c9200f0e7d9aafb31057692f45c202cdcb43a3f9b37c94e7c84" },
{ url = "https://mirrors.aliyun.com/pypi/packages/db/40/22de86e966e648ea0e3e438ad523c86d0cf4866b3841e248726fb4afded8/mysql_connector_python-9.3.0-cp313-cp313-macosx_14_0_x86_64.whl", hash = "sha256:495798dd34445d749991fb3a2aa87b4205100676939556d8d4aab5d5558e7a1f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4c/19/36983937347b6a58af546950c88a9403cdce944893850e80ffb7f602a099/mysql_connector_python-9.3.0-cp313-cp313-manylinux_2_28_aarch64.whl", hash = "sha256:be0ef15f6023ae2037347498f005a4471f694f8a6b8384c3194895e153120286" },
{ url = "https://mirrors.aliyun.com/pypi/packages/18/12/7ccbc678a130df0f751596b37eddb98b2e40930d0ebc9ee41965ffbf0b92/mysql_connector_python-9.3.0-cp313-cp313-manylinux_2_28_x86_64.whl", hash = "sha256:4364d3a37c449f1c0bb9e52fd4eddc620126b9897b6b9f2fd1b3f33dacc16356" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c2/5e/c361caa024ce14ffc1f5b153d90f0febf5e9483a60c4b5c84e1e012363cc/mysql_connector_python-9.3.0-cp313-cp313-win_amd64.whl", hash = "sha256:2a5de57814217077a8672063167b616b1034a37b614b93abcb602cc0b8c6fade" },
{ url = "https://mirrors.aliyun.com/pypi/packages/23/1d/8c2c6672094b538f4881f7714e5332fdcddd05a7e196cbc9eb4a9b5e9a45/mysql_connector_python-9.3.0-py2.py3-none-any.whl", hash = "sha256:8ab7719d614cf5463521082fab86afc21ada504b538166090e00eeaa1ff729bc" },
Refact: update description for max_token in embedding #12792 (#12845) ### What problem does this PR solve? Refact: update description for max_token in embedding #12792 ### Type of change - [x] Refactoring Co-authored-by: Liu An <asiro@qq.com>
2026-01-28 09:52:32 +08:00
]
Feat: add basic Langfuse support for LLM module (#6443) ### What problem does this PR solve? #6155 Add basic Langfuse support for LLM module. A trace example: <img width="755" alt="image" src="https://github.com/user-attachments/assets/25c1f852-5116-486c-a47f-6097187142ca" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-03-24 13:18:47 +08:00
[[package]]
name = "nest-asyncio"
version = "1.6.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/83/f8/51569ac65d696c8ecbee95938f89d4abf00f47d58d48f6fbabfe8f0baefe/nest_asyncio-1.6.0.tar.gz", hash = "sha256:6f172d5449aca15afd6c646851f4e31e02c598d553a667e38cafa997cfec55fe" }
Feat: add basic Langfuse support for LLM module (#6443) ### What problem does this PR solve? #6155 Add basic Langfuse support for LLM module. A trace example: <img width="755" alt="image" src="https://github.com/user-attachments/assets/25c1f852-5116-486c-a47f-6097187142ca" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-03-24 13:18:47 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/a0/c4/c2971a3ba4c6103a3d10c4b0f24f461ddc027f0f09763220cf35ca1401b3/nest_asyncio-1.6.0-py3-none-any.whl", hash = "sha256:87af6efd6b5e897c81050477ef65c62e2b2f35d51703cae01aff2905b1852e1c" },
Feat: add basic Langfuse support for LLM module (#6443) ### What problem does this PR solve? #6155 Add basic Langfuse support for LLM module. A trace example: <img width="755" alt="image" src="https://github.com/user-attachments/assets/25c1f852-5116-486c-a47f-6097187142ca" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-03-24 13:18:47 +08:00
]
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
[[package]]
name = "networkx"
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
version = "3.6.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/6a/51/63fe664f3908c97be9d2e4f1158eb633317598cfa6e1fc14af5383f17512/networkx-3.6.1.tar.gz", hash = "sha256:26b7c357accc0c8cde558ad486283728b65b6a95d85ee1cd66bafab4c8168509" }
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/9e/c9/b2622292ea83fbb4ec318f5b9ab867d0a28ab43c5717bb85b0a5f6b3b0a4/networkx-3.6.1-py3-none-any.whl", hash = "sha256:d47fbf302e7d9cbbb9e2555a0d267983d2aa476bac30e90dfbe5669bd57f3762" },
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "nltk"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "3.9.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "click" },
{ name = "joblib" },
{ name = "regex" },
{ name = "tqdm" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f9/76/3a5e4312c19a028770f86fd7c058cf9f4ec4321c6cf7526bab998a5b683c/nltk-3.9.2.tar.gz", hash = "sha256:0f409e9b069ca4177c1903c3e843eef90c7e92992fa4931ae607da6de49e1419" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/60/90/81ac364ef94209c100e12579629dc92bf7a709a84af32f8c551b02c07e94/nltk-3.9.2-py3-none-any.whl", hash = "sha256:1e209d2b3009110635ed9709a67a1a3e33a10f799490fa71cf4bec218c11c88a" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "numba"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "0.63.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "llvmlite" },
{ name = "numpy" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/dc/60/0145d479b2209bd8fdae5f44201eceb8ce5a23e0ed54c71f57db24618665/numba-0.63.1.tar.gz", hash = "sha256:b320aa675d0e3b17b40364935ea52a7b1c670c9037c39cf92c49502a75902f4b" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/57/f7/e19e6eff445bec52dde5bed1ebb162925a8e6f988164f1ae4b3475a73680/numba-0.63.1-cp313-cp313-macosx_12_0_arm64.whl", hash = "sha256:0bd4fd820ef7442dcc07da184c3f54bb41d2bdb7b35bacf3448e73d081f730dc" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e9/6c/1e222edba1e20e6b113912caa9b1665b5809433cbcb042dfd133c6f1fd38/numba-0.63.1-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:53de693abe4be3bd4dee38e1c55f01c55ff644a6a3696a3670589e6e4c39cde2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/76/0a/590bad11a8b3feeac30a24d01198d46bdb76ad15c70d3a530691ce3cae58/numba-0.63.1-cp313-cp313-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:81227821a72a763c3d4ac290abbb4371d855b59fdf85d5af22a47c0e86bf8c7e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4e/f5/3800384a24eed1e4d524669cdbc0b9b8a628800bb1e90d7bd676e5f22581/numba-0.63.1-cp313-cp313-win_amd64.whl", hash = "sha256:eb227b07c2ac37b09432a9bda5142047a2d1055646e089d4a240a2643e508102" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "numpy"
version = "1.26.4"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/65/6e/09db70a523a96d25e115e71cc56a6f9031e7b8cd166c1ac8438307c14058/numpy-1.26.4.tar.gz", hash = "sha256:2a02aba9ed12e4ac4eb3ea9421c420301a0c6460d9830d74a9df87efa4912010" }
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
[[package]]
name = "oauthlib"
version = "3.3.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/0b/5f/19930f824ffeb0ad4372da4812c50edbd1434f678c90c2733e1188edfc63/oauthlib-3.3.1.tar.gz", hash = "sha256:0f0f8aa759826a193cf66c12ea1af1637f87b9b4622d46e866952bb022e538c9" }
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/be/9c/92789c596b8df838baa98fa71844d84283302f7604ed565dafe5a6b5041a/oauthlib-3.3.1-py3-none-any.whl", hash = "sha256:88119c938d2b8fb88561af5f6ee0eec8cc8d552b7bb1f712743136eb7523b7a1" },
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
]
[[package]]
name = "office365-rest-python-client"
version = "2.6.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
dependencies = [
{ name = "msal" },
{ name = "pytz" },
{ name = "requests" },
{ name = "typing-extensions" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/bc/04/6dce2d581c54a8e55a3b128cf79a93821a68a62bb9a956e65476c5bb247e/office365_rest_python_client-2.6.2.tar.gz", hash = "sha256:ce27f5a1c0cc3ff97041ccd9b386145692be4c64739f243f7d6ac3edbe0a3c46" }
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/3a/a4/611155711f8af347875c15b8b83f5fd9e978bd4de45f90085b9a583b684d/Office365_REST_Python_Client-2.6.2-py3-none-any.whl", hash = "sha256:06fc6829c39b503897caa9d881db419d7f97a8e4f1c95c4c2d12db36ea6c955d" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
[[package]]
name = "olefile"
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
version = "0.47"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/69/1b/077b508e3e500e1629d366249c3ccb32f95e50258b231705c09e3c7a4366/olefile-0.47.zip", hash = "sha256:599383381a0bf3dfbd932ca0ca6515acd174ed48870cbf7fee123d698c192c1c" }
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/17/d3/b64c356a907242d719fc668b71befd73324e47ab46c8ebbbede252c154b2/olefile-0.47-py2.py3-none-any.whl", hash = "sha256:543c7da2a7adadf21214938bb79c83ea12b473a4b6ee4ad4bf854e7715e13d1f" },
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
]
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
[[package]]
name = "oletools"
version = "0.60.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
dependencies = [
{ name = "colorclass" },
{ name = "easygui" },
{ name = "msoffcrypto-tool", marker = "(platform_python_implementation != 'PyPy' and sys_platform == 'darwin') or (platform_python_implementation != 'PyPy' and sys_platform == 'win32') or (sys_platform != 'darwin' and sys_platform != 'win32')" },
{ name = "olefile" },
{ name = "pcodedmp" },
{ name = "pyparsing" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/5c/2f/037f40e44706d542b94a2312ccc33ee2701ebfc9a83b46b55263d49ce55a/oletools-0.60.2.zip", hash = "sha256:ad452099f4695ffd8855113f453348200d195ee9fa341a09e197d66ee7e0b2c3" }
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ac/ff/05257b7183279b80ecec6333744de23f48f0faeeba46c93e6d13ce835515/oletools-0.60.2-py2.py3-none-any.whl", hash = "sha256:72ad8bd748fd0c4e7b5b4733af770d11543ebb2bf2697455f99f975fcd50cc96" },
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "ollama"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "0.6.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "httpx" },
Fix problem with Google Cloud models with reasoning (like gemini) - Additional fix to issue #10474 (#10502) ### What problem does this PR solve? Issue #10474 - Update to PR #10477 ### Type of change - [X] Bug Fix (non-breaking change which fixes an issue)
2025-10-15 08:54:20 +02:00
{ name = "pydantic" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/9d/5a/652dac4b7affc2b37b95386f8ae78f22808af09d720689e3d7a86b6ed98e/ollama-0.6.1.tar.gz", hash = "sha256:478c67546836430034b415ed64fa890fd3d1ff91781a9d548b3325274e69d7c6" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/47/4f/4a617ee93d8208d2bcf26b2d8b9402ceaed03e3853c754940e2290fed063/ollama-0.6.1-py3-none-any.whl", hash = "sha256:fc4c984b345735c5486faeee67d8a265214a31cbb828167782dc642ce0a2bf8c" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "onnxruntime"
Bump python to >=3.12 (#11846) ### What problem does this PR solve? Bump python to >=3.12 ### Type of change - [x] Refactoring
2025-12-09 19:55:25 +08:00
version = "1.23.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
Fix onnxruntime-gpu marks (#4643) ### What problem does this PR solve? Fix onnxruntime-gpu marks ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-01-26 09:37:59 +08:00
{ name = "coloredlogs" },
{ name = "flatbuffers" },
{ name = "numpy" },
{ name = "packaging" },
{ name = "protobuf" },
{ name = "sympy" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/3d/41/fba0cabccecefe4a1b5fc8020c44febb334637f133acefc7ec492029dd2c/onnxruntime-1.23.2-cp313-cp313-macosx_13_0_arm64.whl", hash = "sha256:2ff531ad8496281b4297f32b83b01cdd719617e2351ffe0dba5684fb283afa1f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fe/f9/2d49ca491c6a986acce9f1d1d5fc2099108958cc1710c28e89a032c9cfe9/onnxruntime-1.23.2-cp313-cp313-macosx_13_0_x86_64.whl", hash = "sha256:162f4ca894ec3de1a6fd53589e511e06ecdc3ff646849b62a9da7489dee9ce95" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1c/a1/428ee29c6eaf09a6f6be56f836213f104618fb35ac6cc586ff0f477263eb/onnxruntime-1.23.2-cp313-cp313-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:45d127d6e1e9b99d1ebeae9bcd8f98617a812f53f46699eafeb976275744826b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f2/2b/b57c8a2466a3126dbe0a792f56ad7290949b02f47b86216cd47d857e4b77/onnxruntime-1.23.2-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:8bace4e0d46480fbeeb7bbe1ffe1f080e6663a42d1086ff95c1551f2d39e7872" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4a/93/aba75358133b3a941d736816dd392f687e7eab77215a6e429879080b76b6/onnxruntime-1.23.2-cp313-cp313-win_amd64.whl", hash = "sha256:1f9cc0a55349c584f083c1c076e611a7c35d5b867d5d6e6d6c823bf821978088" },
{ url = "https://mirrors.aliyun.com/pypi/packages/7c/3d/6830fa61c69ca8e905f237001dbfc01689a4e4ab06147020a4518318881f/onnxruntime-1.23.2-cp313-cp313t-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:9d2385e774f46ac38f02b3a91a91e30263d41b2f1f4f26ae34805b2a9ddef466" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b6/ca/862b1e7a639460f0ca25fd5b6135fb42cf9deea86d398a92e44dfda2279d/onnxruntime-1.23.2-cp313-cp313t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:e2b9233c4947907fd1818d0e581c049c41ccc39b2856cc942ff6d26317cee145" },
Added onnxruntime-gpu (#4631) ### What problem does this PR solve? Added onnxruntime-gpu ### Type of change - [x] Refactoring
2025-01-24 14:33:21 +08:00
]
[[package]]
name = "onnxruntime-gpu"
Bump python to >=3.12 (#11846) ### What problem does this PR solve? Bump python to >=3.12 ### Type of change - [x] Refactoring
2025-12-09 19:55:25 +08:00
version = "1.23.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Added onnxruntime-gpu (#4631) ### What problem does this PR solve? Added onnxruntime-gpu ### Type of change - [x] Refactoring
2025-01-24 14:33:21 +08:00
dependencies = [
Fix: upgrade pyasn1 from 0.6.2 to 0.6.3 to address CVE-2026-30922 (#13773) ## Summary - Adds `pyasn1>=0.6.3` as a `[tool.uv.constraint-dependencies]` entry to mitigate **CVE-2026-30922** (CVSS 7.5 HIGH) - Regenerates `uv.lock` so the resolved pyasn1 version moves from **0.6.2 to 0.6.3** ## Details **CVE-2026-30922** is a Denial of Service vulnerability in pyasn1 caused by unbounded recursion when decoding ASN.1 data with deeply nested structures. An attacker can send crafted payloads with thousands of nested SEQUENCE or SET tags to trigger a `RecursionError` crash or memory exhaustion. - **Severity:** HIGH (CVSS 7.5) - **Affected versions:** pyasn1 < 0.6.3 - **Fixed in:** pyasn1 >= 0.6.3 - **NVD:** https://nvd.nist.gov/vuln/detail/CVE-2026-25769 `pyasn1` is not a direct dependency of RAGFlow but is pulled in transitively via `google-auth` -> `rsa` -> `pyasn1-modules` -> `pyasn1`. The `constraint-dependencies` mechanism in uv is the correct way to enforce a minimum version for transitive dependencies without polluting the direct dependency list. ## Test plan - [x] `pyproject.toml` passes TOML validation - [x] `uv lock` resolves successfully with the new constraint - [x] pyasn1 version in `uv.lock` is now 0.6.3 - [ ] Existing CI/CD tests continue to pass Closes #13686
2026-03-26 19:37:34 -07:00
{ name = "coloredlogs", marker = "(platform_machine != 'aarch64' and sys_platform == 'linux') or (sys_platform != 'darwin' and sys_platform != 'linux')" },
{ name = "flatbuffers", marker = "(platform_machine != 'aarch64' and sys_platform == 'linux') or (sys_platform != 'darwin' and sys_platform != 'linux')" },
{ name = "numpy", marker = "(platform_machine != 'aarch64' and sys_platform == 'linux') or (sys_platform != 'darwin' and sys_platform != 'linux')" },
{ name = "packaging", marker = "(platform_machine != 'aarch64' and sys_platform == 'linux') or (sys_platform != 'darwin' and sys_platform != 'linux')" },
{ name = "protobuf", marker = "(platform_machine != 'aarch64' and sys_platform == 'linux') or (sys_platform != 'darwin' and sys_platform != 'linux')" },
{ name = "sympy", marker = "(platform_machine != 'aarch64' and sys_platform == 'linux') or (sys_platform != 'darwin' and sys_platform != 'linux')" },
Added onnxruntime-gpu (#4631) ### What problem does this PR solve? Added onnxruntime-gpu ### Type of change - [x] Refactoring
2025-01-24 14:33:21 +08:00
]
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/03/05/40d561636e4114b54aa06d2371bfbca2d03e12cfdf5d4b85814802f18a75/onnxruntime_gpu-1.23.2-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:1e8f75af5da07329d0c3a5006087f4051d8abd133b4be7c9bae8cdab7bea4c26" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b6/3b/418300438063d403384c79eaef1cb13c97627042f2247b35a887276a355a/onnxruntime_gpu-1.23.2-cp313-cp313-win_amd64.whl", hash = "sha256:7f1b3f49e5e126b99e23ec86b4203db41c2a911f6165f7624f2bc8267aaca767" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b8/dc/80b145e3134d7eba31309b3299a2836e37c76e4c419a261ad9796f8f8d65/onnxruntime_gpu-1.23.2-cp313-cp313t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:20959cd4ae358aab6579ab9123284a7b1498f7d51ec291d429a5edc26511306f" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "openai"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "2.41.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "anyio" },
{ name = "distro" },
{ name = "httpx" },
{ name = "jiter" },
{ name = "pydantic" },
{ name = "sniffio" },
{ name = "tqdm" },
{ name = "typing-extensions" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/3c/a6/5815fe2e2aca74b36c650d1bd43b69827cee568073d0d2d9b6fc5aaac80c/openai-2.41.0.tar.gz", hash = "sha256:db5c362acd6604b84f076abbefa66826ea4b46ecba2954ed866e6a149a1352c0" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/be/51/d82bb424e8aa372190c5233253a2ceb399a778747d18b42cff487411e663/openai-2.41.0-py3-none-any.whl", hash = "sha256:20cc7952e8501c7e5773dd2ef7be437bae9cb549044902e1041a83a54516e375" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "opencv-python"
version = "4.10.0.84"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "numpy" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/4a/e7/b70a2d9ab205110d715906fc8ec83fbb00404aeb3a37a0654fdb68eb0c8c/opencv-python-4.10.0.84.tar.gz", hash = "sha256:72d234e4582e9658ffea8e9cae5b63d488ad06994ef12d81dc303b17472f3526" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/66/82/564168a349148298aca281e342551404ef5521f33fba17b388ead0a84dc5/opencv_python-4.10.0.84-cp37-abi3-macosx_11_0_arm64.whl", hash = "sha256:fc182f8f4cda51b45f01c64e4cbedfc2f00aff799debebc305d8d0210c43f251" },
{ url = "https://mirrors.aliyun.com/pypi/packages/64/4a/016cda9ad7cf18c58ba074628a4eaae8aa55f3fd06a266398cef8831a5b9/opencv_python-4.10.0.84-cp37-abi3-macosx_12_0_x86_64.whl", hash = "sha256:71e575744f1d23f79741450254660442785f45a0797212852ee5199ef12eed98" },
{ url = "https://mirrors.aliyun.com/pypi/packages/81/e4/7a987ebecfe5ceaf32db413b67ff18eb3092c598408862fff4d7cc3fd19b/opencv_python-4.10.0.84-cp37-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:09a332b50488e2dda866a6c5573ee192fe3583239fb26ff2f7f9ceb0bc119ea6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3f/a4/d2537f47fd7fcfba966bd806e3ec18e7ee1681056d4b0a9c8d983983e4d5/opencv_python-4.10.0.84-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9ace140fc6d647fbe1c692bcb2abce768973491222c067c131d80957c595b71f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1e/39/bbf57e7b9dab623e8773f6ff36385456b7ae7fa9357a5e53db732c347eac/opencv_python-4.10.0.84-cp37-abi3-win32.whl", hash = "sha256:2db02bb7e50b703f0a2d50c50ced72e95c574e1e5a0bb35a8a86d0b35c98c236" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ec/6c/fab8113424af5049f85717e8e527ca3773299a3c6b02506e66436e19874f/opencv_python-4.10.0.84-cp37-abi3-win_amd64.whl", hash = "sha256:32dbbd94c26f611dc5cc6979e6b7aa1f55a64d6b463cc1dcd3c95505a63e48fe" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "opencv-python-headless"
version = "4.10.0.84"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "numpy" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/2f/7e/d20f68a5f1487adf19d74378d349932a386b1ece3be9be9915e5986db468/opencv-python-headless-4.10.0.84.tar.gz", hash = "sha256:f2017c6101d7c2ef8d7bc3b414c37ff7f54d64413a1847d89970b6b7069b4e1a" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/1c/9b/583c8d9259f6fc19413f83fd18dd8e6cbc8eefb0b4dc6da52dd151fe3272/opencv_python_headless-4.10.0.84-cp37-abi3-macosx_11_0_arm64.whl", hash = "sha256:a4f4bcb07d8f8a7704d9c8564c224c8b064c63f430e95b61ac0bffaa374d330e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c0/7b/b4c67f5dad7a9a61c47f7a39e4050e8a4628bd64b3c3daaeb755d759f928/opencv_python_headless-4.10.0.84-cp37-abi3-macosx_12_0_x86_64.whl", hash = "sha256:5ae454ebac0eb0a0b932e3406370aaf4212e6a3fdb5038cc86c7aea15a6851da" },
{ url = "https://mirrors.aliyun.com/pypi/packages/91/61/f838ce2046f3ec3591ea59ea3549085e399525d3b4558c4ed60b55ed88c0/opencv_python_headless-4.10.0.84-cp37-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:46071015ff9ab40fccd8a163da0ee14ce9846349f06c6c8c0f2870856ffa45db" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d1/09/248f86a404567303cdf120e4a301f389b68e3b18e5c0cc428de327da609c/opencv_python_headless-4.10.0.84-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:377d08a7e48a1405b5e84afcbe4798464ce7ee17081c1c23619c8b398ff18295" },
{ url = "https://mirrors.aliyun.com/pypi/packages/30/c0/66f88d58500e990a9a0a5c06f98862edf1d0a3a430781218a8c193948438/opencv_python_headless-4.10.0.84-cp37-abi3-win32.whl", hash = "sha256:9092404b65458ed87ce932f613ffbb1106ed2c843577501e5768912360fc50ec" },
{ url = "https://mirrors.aliyun.com/pypi/packages/26/d0/22f68eb23eea053a31655960f133c0be9726c6a881547e6e9e7e2a946c4f/opencv_python_headless-4.10.0.84-cp37-abi3-win_amd64.whl", hash = "sha256:afcf28bd1209dd58810d33defb622b325d3cbe49dcd7a43a902982c33e5fad05" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Oss support opendal(including mysql) (#8204) ### What problem does this PR solve? #8074 Oss support opendal(including mysql) ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2025-06-12 11:37:42 +08:00
[[package]]
name = "opendal"
version = "0.45.20"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/2f/3f/927dfe1349ae58b9238b8eafba747af648d660a9425f486dda01a10f0b78/opendal-0.45.20.tar.gz", hash = "sha256:9f6f90d9e9f9d6e9e5a34aa7729169ef34d2f1869ad1e01ddc39b1c0ce0c9405" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/84/77/6427e16b8630f0cc71f4a1b01648ed3264f1e04f1f6d9b5d09e5c6a4dd2f/opendal-0.45.20-cp311-abi3-macosx_10_12_x86_64.macosx_11_0_arm64.macosx_10_12_universal2.whl", hash = "sha256:35acdd8001e4a741532834fdbff3020ffb10b40028bb49fbe93c4f8197d66d8c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/12/1f/83e415334739f1ab4dba55cdd349abf0b66612249055afb422a354b96ac8/opendal-0.45.20-cp311-abi3-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:629bfe8d384364bced6cbeb01f49b99779fa5151c68048a1869ff645ddcfcb25" },
{ url = "https://mirrors.aliyun.com/pypi/packages/49/94/c5de6ed54a02d7413636c2ccefa71d8dd09c2ada1cd6ecab202feb1fdeda/opendal-0.45.20-cp311-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d12cc5ac7e441fb93d86d1673112d9fb08580fc3226f864434f4a56a72efec53" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c6/83/713a1e1de8cbbd69af50e26644bbdeef3c1068b89f442417376fa3c0f591/opendal-0.45.20-cp311-abi3-manylinux_2_28_aarch64.whl", hash = "sha256:45a3adae1f473052234fc4054a6f210df3ded9aff10db8d545d0a37eff3b13cc" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c7/78/c9651e753aaf6eb61887ca372a3f9c2ae57dae03c3159d24deaf018c26dc/opendal-0.45.20-cp311-abi3-musllinux_1_1_aarch64.whl", hash = "sha256:d8947857052c85a4b0e251d50e23f5f68f0cdd9e509e32e614a5e4b2fc7424c4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3c/9d/5d8c20c0fc93df5e349e5694167de30afdc54c5755704cc64764a6cbb309/opendal-0.45.20-cp311-abi3-musllinux_1_1_armv7l.whl", hash = "sha256:891d2f9114efeef648973049ed15e56477e8feb9e48b540bd8d6105ea22a253c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/21/39/05262f748a2085522e0c85f03eab945589313dc9caedc002872c39162776/opendal-0.45.20-cp311-abi3-musllinux_1_1_x86_64.whl", hash = "sha256:539de9b825f6783d6289d88c0c9ac5415daa4d892d761e3540c565bda51e8997" },
{ url = "https://mirrors.aliyun.com/pypi/packages/74/83/cc7c6de29b0a7585cd445258d174ca204d37729c3874ad08e515b0bf331c/opendal-0.45.20-cp311-abi3-win_amd64.whl", hash = "sha256:145efd56aa33b493d5b652c3e4f5ae5097ab69d38c132d80f108e9f5c1e4d863" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ab/a9/7a4ae9b309c5a675cfedd22c1d020b4f7d3a69a7a4db104d327391a0bf95/opendal-0.45.20-cp313-cp313t-macosx_10_12_x86_64.macosx_11_0_arm64.macosx_10_12_universal2.whl", hash = "sha256:1170ea1f3f082b9ccffe49b2483ddcccbf7cac1f3d2398391f7702e15ab5286a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/95/61/3de0cbdcf95b7164e5701d83f621bb85ca3dd7a8c9835476a4904121d5a1/opendal-0.45.20-cp313-cp313t-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:7afc556fca1442f7c667bdebe38ab77d87bae7fa3dd9ac8a42525fd110b12095" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c9/de/1ac5db8e66817d7526af2ecac6d7afca6949322d6a208dd436e87ea08007/opendal-0.45.20-cp313-cp313t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a8eb1da41f7d729ca0639e4ec6481515ad5b94ef016b0d5cf9e34b391ca0086c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/bb/36/df5fe4db6bef0e008238df573227d84dfb1d57c03717dcc400920d96e2e7/opendal-0.45.20-cp313-cp313t-manylinux_2_28_aarch64.whl", hash = "sha256:b19a4428a2a5234f6dab569b36a1423f67ff243aafc36ab1958f709304a2b580" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a1/ed/3b00e7ecef6fee7ce335aa5da2077855df6f2e1eac40d7f4a98a5de5f5e7/opendal-0.45.20-cp313-cp313t-musllinux_1_1_aarch64.whl", hash = "sha256:1d6fc2e24cdcb26c86d5db213f8034db41386a25343b70dca01e79874f849e4c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0c/c4/e2e89ac6d63f412dad51243184270f5095eb7c6ee00504b245df65957efb/opendal-0.45.20-cp313-cp313t-musllinux_1_1_armv7l.whl", hash = "sha256:ccf9a1cab13fa4f4c44a14f85fe42ec99ba2595808fae44d753ef2f9ba7b55e3" },
{ url = "https://mirrors.aliyun.com/pypi/packages/39/29/b02073b097915eb4c381e21bb48e5ae16dd5a6b6edb5995fb87607a82b0e/opendal-0.45.20-cp313-cp313t-musllinux_1_1_x86_64.whl", hash = "sha256:197239aa43221c8bc9b46cfeeeafbfd2e59eee7a0ec4cbe83d89a3efd53c24a3" },
{ url = "https://mirrors.aliyun.com/pypi/packages/01/ec/d149ed82a5cc175460e044e040d2e09e496c74e699112c6ee9d1828ff6a4/opendal-0.45.20-cp313-cp313t-win_amd64.whl", hash = "sha256:5af03824ffca796a2c77b570760bb7ddc754e9485f882fed5cc834aab4772cbf" },
Oss support opendal(including mysql) (#8204) ### What problem does this PR solve? #8074 Oss support opendal(including mysql) ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2025-06-12 11:37:42 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "openpyxl"
version = "3.1.5"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "et-xmlfile" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/3d/f9/88d94a75de065ea32619465d2f77b29a0469500e99012523b91cc4141cd1/openpyxl-3.1.5.tar.gz", hash = "sha256:cf0e3cf56142039133628b5acffe8ef0c12bc902d2aadd3e0fe5878dc08d1050" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/c0/da/977ded879c29cbd04de313843e76868e6e13408a94ed6b987245dc7c8506/openpyxl-3.1.5-py2.py3-none-any.whl", hash = "sha256:5282c12b107bffeef825f4617dc029afaf41d0ea60823bbb665ef3079dc79de2" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: Adds OpenSearch2.19.1 as the vector_database support (#7140) ### What problem does this PR solve? This PR adds the support for latest OpenSearch2.19.1 as the store engine & search engine option for RAGFlow. ### Main Benefit 1. OpenSearch2.19.1 is licensed under the [Apache v2.0 License] which is much better than Elasticsearch 2. For search, OpenSearch2.19.1 supports full-text search、vector_search、hybrid_search those are similar with Elasticsearch on schema 3. For store, OpenSearch2.19.1 stores text、vector those are quite simliar with Elasticsearch on schema ### Changes - Support opensearch_python_connetor. I make a lot of adaptions since the schema and api/method between ES and Opensearch differs in many ways(especially the knn_search has a significant gap) : rag/utils/opensearch_coon.py - Support static config adaptions by changing: conf/service_conf.yaml、api/settings.py、rag/settings.py - Supprt some store&search schema changes between OpenSearch and ES: conf/os_mapping.json - Support OpenSearch python sdk : pyproject.toml - Support docker config for OpenSearch2.19.1 : docker/.env、docker/docker-compose-base.yml、docker/service_conf.yaml.template ### How to use - I didn't change the priority that ES as the default doc/search engine. Only if in docker/.env , we set DOC_ENGINE=${DOC_ENGINE:-opensearch}, it will work. ### Others Our team tested a lot of docs in our environment by using OpenSearch as the vector database ,it works very well. All the conifg for OpenSearch is necessary. ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Yongteng Lei <yongtengrey@outlook.com> Co-authored-by: writinwaters <93570324+writinwaters@users.noreply.github.com> Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
2025-04-24 16:03:31 +08:00
[[package]]
name = "opensearch-py"
version = "2.7.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Adds OpenSearch2.19.1 as the vector_database support (#7140) ### What problem does this PR solve? This PR adds the support for latest OpenSearch2.19.1 as the store engine & search engine option for RAGFlow. ### Main Benefit 1. OpenSearch2.19.1 is licensed under the [Apache v2.0 License] which is much better than Elasticsearch 2. For search, OpenSearch2.19.1 supports full-text search、vector_search、hybrid_search those are similar with Elasticsearch on schema 3. For store, OpenSearch2.19.1 stores text、vector those are quite simliar with Elasticsearch on schema ### Changes - Support opensearch_python_connetor. I make a lot of adaptions since the schema and api/method between ES and Opensearch differs in many ways(especially the knn_search has a significant gap) : rag/utils/opensearch_coon.py - Support static config adaptions by changing: conf/service_conf.yaml、api/settings.py、rag/settings.py - Supprt some store&search schema changes between OpenSearch and ES: conf/os_mapping.json - Support OpenSearch python sdk : pyproject.toml - Support docker config for OpenSearch2.19.1 : docker/.env、docker/docker-compose-base.yml、docker/service_conf.yaml.template ### How to use - I didn't change the priority that ES as the default doc/search engine. Only if in docker/.env , we set DOC_ENGINE=${DOC_ENGINE:-opensearch}, it will work. ### Others Our team tested a lot of docs in our environment by using OpenSearch as the vector database ,it works very well. All the conifg for OpenSearch is necessary. ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Yongteng Lei <yongtengrey@outlook.com> Co-authored-by: writinwaters <93570324+writinwaters@users.noreply.github.com> Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
2025-04-24 16:03:31 +08:00
dependencies = [
{ name = "certifi" },
{ name = "events" },
{ name = "python-dateutil" },
{ name = "requests" },
{ name = "urllib3" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/c4/ca/5be52de5c69ecd327c16f3fc0dba82b7ffda5bbd0c0e215bdf23a4d12b12/opensearch_py-2.7.1.tar.gz", hash = "sha256:67ab76e9373669bc71da417096df59827c08369ac3795d5438c9a8be21cbd759" }
Feat: Adds OpenSearch2.19.1 as the vector_database support (#7140) ### What problem does this PR solve? This PR adds the support for latest OpenSearch2.19.1 as the store engine & search engine option for RAGFlow. ### Main Benefit 1. OpenSearch2.19.1 is licensed under the [Apache v2.0 License] which is much better than Elasticsearch 2. For search, OpenSearch2.19.1 supports full-text search、vector_search、hybrid_search those are similar with Elasticsearch on schema 3. For store, OpenSearch2.19.1 stores text、vector those are quite simliar with Elasticsearch on schema ### Changes - Support opensearch_python_connetor. I make a lot of adaptions since the schema and api/method between ES and Opensearch differs in many ways(especially the knn_search has a significant gap) : rag/utils/opensearch_coon.py - Support static config adaptions by changing: conf/service_conf.yaml、api/settings.py、rag/settings.py - Supprt some store&search schema changes between OpenSearch and ES: conf/os_mapping.json - Support OpenSearch python sdk : pyproject.toml - Support docker config for OpenSearch2.19.1 : docker/.env、docker/docker-compose-base.yml、docker/service_conf.yaml.template ### How to use - I didn't change the priority that ES as the default doc/search engine. Only if in docker/.env , we set DOC_ENGINE=${DOC_ENGINE:-opensearch}, it will work. ### Others Our team tested a lot of docs in our environment by using OpenSearch as the vector database ,it works very well. All the conifg for OpenSearch is necessary. ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Yongteng Lei <yongtengrey@outlook.com> Co-authored-by: writinwaters <93570324+writinwaters@users.noreply.github.com> Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
2025-04-24 16:03:31 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/80/8f/db678ae203d761922a73920215ea53a79faf3bb1ec6aa9511f809c8e234c/opensearch_py-2.7.1-py3-none-any.whl", hash = "sha256:5417650eba98a1c7648e502207cebf3a12beab623ffe0ebbf55f9b1b4b6e44e9" },
Feat: Adds OpenSearch2.19.1 as the vector_database support (#7140) ### What problem does this PR solve? This PR adds the support for latest OpenSearch2.19.1 as the store engine & search engine option for RAGFlow. ### Main Benefit 1. OpenSearch2.19.1 is licensed under the [Apache v2.0 License] which is much better than Elasticsearch 2. For search, OpenSearch2.19.1 supports full-text search、vector_search、hybrid_search those are similar with Elasticsearch on schema 3. For store, OpenSearch2.19.1 stores text、vector those are quite simliar with Elasticsearch on schema ### Changes - Support opensearch_python_connetor. I make a lot of adaptions since the schema and api/method between ES and Opensearch differs in many ways(especially the knn_search has a significant gap) : rag/utils/opensearch_coon.py - Support static config adaptions by changing: conf/service_conf.yaml、api/settings.py、rag/settings.py - Supprt some store&search schema changes between OpenSearch and ES: conf/os_mapping.json - Support OpenSearch python sdk : pyproject.toml - Support docker config for OpenSearch2.19.1 : docker/.env、docker/docker-compose-base.yml、docker/service_conf.yaml.template ### How to use - I didn't change the priority that ES as the default doc/search engine. Only if in docker/.env , we set DOC_ENGINE=${DOC_ENGINE:-opensearch}, it will work. ### Others Our team tested a lot of docs in our environment by using OpenSearch as the vector database ,it works very well. All the conifg for OpenSearch is necessary. ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Yongteng Lei <yongtengrey@outlook.com> Co-authored-by: writinwaters <93570324+writinwaters@users.noreply.github.com> Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
2025-04-24 16:03:31 +08:00
]
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
[[package]]
name = "opentelemetry-api"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.39.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
dependencies = [
{ name = "importlib-metadata" },
{ name = "typing-extensions" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/97/b9/3161be15bb8e3ad01be8be5a968a9237c3027c5be504362ff800fca3e442/opentelemetry_api-1.39.1.tar.gz", hash = "sha256:fbde8c80e1b937a2c61f20347e91c0c18a1940cecf012d62e65a7caf08967c9c" }
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/cf/df/d3f1ddf4bb4cb50ed9b1139cc7b1c54c34a1e7ce8fd1b9a37c0d1551a6bd/opentelemetry_api-1.39.1-py3-none-any.whl", hash = "sha256:2edd8463432a7f8443edce90972169b195e7d6a05500cd29e6d13898187c9950" },
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
]
[[package]]
name = "opentelemetry-exporter-otlp-proto-common"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.39.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
dependencies = [
{ name = "opentelemetry-proto" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/e9/9d/22d241b66f7bbde88a3bfa6847a351d2c46b84de23e71222c6aae25c7050/opentelemetry_exporter_otlp_proto_common-1.39.1.tar.gz", hash = "sha256:763370d4737a59741c89a67b50f9e39271639ee4afc999dadfe768541c027464" }
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/8c/02/ffc3e143d89a27ac21fd557365b98bd0653b98de8a101151d5805b5d4c33/opentelemetry_exporter_otlp_proto_common-1.39.1-py3-none-any.whl", hash = "sha256:08f8a5862d64cc3435105686d0216c1365dc5701f86844a8cd56597d0c764fde" },
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
]
[[package]]
name = "opentelemetry-exporter-otlp-proto-http"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.39.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
dependencies = [
{ name = "googleapis-common-protos" },
{ name = "opentelemetry-api" },
{ name = "opentelemetry-exporter-otlp-proto-common" },
{ name = "opentelemetry-proto" },
{ name = "opentelemetry-sdk" },
{ name = "requests" },
{ name = "typing-extensions" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/80/04/2a08fa9c0214ae38880df01e8bfae12b067ec0793446578575e5080d6545/opentelemetry_exporter_otlp_proto_http-1.39.1.tar.gz", hash = "sha256:31bdab9745c709ce90a49a0624c2bd445d31a28ba34275951a6a362d16a0b9cb" }
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/95/f1/b27d3e2e003cd9a3592c43d099d2ed8d0a947c15281bf8463a256db0b46c/opentelemetry_exporter_otlp_proto_http-1.39.1-py3-none-any.whl", hash = "sha256:d9f5207183dd752a412c4cd564ca8875ececba13be6e9c6c370ffb752fd59985" },
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
]
[[package]]
name = "opentelemetry-proto"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.39.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
dependencies = [
{ name = "protobuf" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/49/1d/f25d76d8260c156c40c97c9ed4511ec0f9ce353f8108ca6e7561f82a06b2/opentelemetry_proto-1.39.1.tar.gz", hash = "sha256:6c8e05144fc0d3ed4d22c2289c6b126e03bcd0e6a7da0f16cedd2e1c2772e2c8" }
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/51/95/b40c96a7b5203005a0b03d8ce8cd212ff23f1793d5ba289c87a097571b18/opentelemetry_proto-1.39.1-py3-none-any.whl", hash = "sha256:22cdc78efd3b3765d09e68bfbd010d4fc254c9818afd0b6b423387d9dee46007" },
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
]
[[package]]
name = "opentelemetry-sdk"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.39.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
dependencies = [
{ name = "opentelemetry-api" },
{ name = "opentelemetry-semantic-conventions" },
{ name = "typing-extensions" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/eb/fb/c76080c9ba07e1e8235d24cdcc4d125ef7aa3edf23eb4e497c2e50889adc/opentelemetry_sdk-1.39.1.tar.gz", hash = "sha256:cf4d4563caf7bff906c9f7967e2be22d0d6b349b908be0d90fb21c8e9c995cc6" }
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/7c/98/e91cf858f203d86f4eccdf763dcf01cf03f1dae80c3750f7e635bfa206b6/opentelemetry_sdk-1.39.1-py3-none-any.whl", hash = "sha256:4d5482c478513ecb0a5d938dcc61394e647066e0cc2676bee9f3af3f3f45f01c" },
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
]
[[package]]
name = "opentelemetry-semantic-conventions"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "0.60b1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
dependencies = [
{ name = "opentelemetry-api" },
{ name = "typing-extensions" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/91/df/553f93ed38bf22f4b999d9be9c185adb558982214f33eae539d3b5cd0858/opentelemetry_semantic_conventions-0.60b1.tar.gz", hash = "sha256:87c228b5a0669b748c76d76df6c364c369c28f1c465e50f661e39737e84bc953" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/7a/5e/5958555e09635d09b75de3c4f8b9cae7335ca545d77392ffe7331534c402/opentelemetry_semantic_conventions-0.60b1-py3-none-any.whl", hash = "sha256:9fa8c8b0c110da289809292b0591220d3a7b53c1526a23021e977d68597893fb" },
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
]
[[package]]
name = "optionaldict"
version = "0.1.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ad/8b/cb72b1d40d21963ad1923ce529ae1f9ba49a13c18c7e6c47081c3dc1211f/optionaldict-0.1.2.tar.gz", hash = "sha256:175b62ee8259af703def42e0b44b413ea55e7b0732456cb2fd5cbbc998605966" }
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/93/c2/408eb8d1ddacc36c6c31e0b0640b8493a4b3e88e26670bdd2b3a8ad7d704/optionaldict-0.1.2-py2.py3-none-any.whl", hash = "sha256:278d9b79b3d6a55b1b8ff5f9017bff6383a33426348d09e699ff50162bbcdab7" },
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "orjson"
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
version = "3.10.18"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/81/0b/fea456a3ffe74e70ba30e01ec183a9b26bec4d497f61dcfce1b601059c60/orjson-3.10.18.tar.gz", hash = "sha256:e8da3947d92123eda795b68228cafe2724815621fe35e8e320a9e9593a4bcd53" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/04/f0/8aedb6574b68096f3be8f74c0b56d36fd94bcf47e6c7ed47a7bd1474aaa8/orjson-3.10.18-cp313-cp313-macosx_10_15_x86_64.macosx_11_0_arm64.macosx_10_15_universal2.whl", hash = "sha256:69c34b9441b863175cc6a01f2935de994025e773f814412030f269da4f7be147" },
{ url = "https://mirrors.aliyun.com/pypi/packages/bc/f7/7118f965541aeac6844fcb18d6988e111ac0d349c9b80cda53583e758908/orjson-3.10.18-cp313-cp313-macosx_15_0_arm64.whl", hash = "sha256:1ebeda919725f9dbdb269f59bc94f861afbe2a27dce5608cdba2d92772364d1c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fb/d9/839637cc06eaf528dd8127b36004247bf56e064501f68df9ee6fd56a88ee/orjson-3.10.18-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:5adf5f4eed520a4959d29ea80192fa626ab9a20b2ea13f8f6dc58644f6927103" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2b/6d/f226ecfef31a1f0e7d6bf9a31a0bbaf384c7cbe3fce49cc9c2acc51f902a/orjson-3.10.18-cp313-cp313-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:7592bb48a214e18cd670974f289520f12b7aed1fa0b2e2616b8ed9e069e08595" },
{ url = "https://mirrors.aliyun.com/pypi/packages/73/2d/371513d04143c85b681cf8f3bce743656eb5b640cb1f461dad750ac4b4d4/orjson-3.10.18-cp313-cp313-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:f872bef9f042734110642b7a11937440797ace8c87527de25e0c53558b579ccc" },
{ url = "https://mirrors.aliyun.com/pypi/packages/69/cb/a4d37a30507b7a59bdc484e4a3253c8141bf756d4e13fcc1da760a0b00cb/orjson-3.10.18-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:0315317601149c244cb3ecef246ef5861a64824ccbcb8018d32c66a60a84ffbc" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1e/ae/cd10883c48d912d216d541eb3db8b2433415fde67f620afe6f311f5cd2ca/orjson-3.10.18-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:e0da26957e77e9e55a6c2ce2e7182a36a6f6b180ab7189315cb0995ec362e049" },
{ url = "https://mirrors.aliyun.com/pypi/packages/6d/4c/2bda09855c6b5f2c055034c9eda1529967b042ff8d81a05005115c4e6772/orjson-3.10.18-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:bb70d489bc79b7519e5803e2cc4c72343c9dc1154258adf2f8925d0b60da7c58" },
{ url = "https://mirrors.aliyun.com/pypi/packages/13/4a/35971fd809a8896731930a80dfff0b8ff48eeb5d8b57bb4d0d525160017f/orjson-3.10.18-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:e9e86a6af31b92299b00736c89caf63816f70a4001e750bda179e15564d7a034" },
{ url = "https://mirrors.aliyun.com/pypi/packages/99/70/0fa9e6310cda98365629182486ff37a1c6578e34c33992df271a476ea1cd/orjson-3.10.18-cp313-cp313-musllinux_1_2_armv7l.whl", hash = "sha256:c382a5c0b5931a5fc5405053d36c1ce3fd561694738626c77ae0b1dfc0242ca1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/32/cb/990a0e88498babddb74fb97855ae4fbd22a82960e9b06eab5775cac435da/orjson-3.10.18-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:8e4b2ae732431127171b875cb2668f883e1234711d3c147ffd69fe5be51a8012" },
{ url = "https://mirrors.aliyun.com/pypi/packages/92/44/473248c3305bf782a384ed50dd8bc2d3cde1543d107138fd99b707480ca1/orjson-3.10.18-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:2d808e34ddb24fc29a4d4041dcfafbae13e129c93509b847b14432717d94b44f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ad/fd/7f1d3edd4ffcd944a6a40e9f88af2197b619c931ac4d3cfba4798d4d3815/orjson-3.10.18-cp313-cp313-win32.whl", hash = "sha256:ad8eacbb5d904d5591f27dee4031e2c1db43d559edb8f91778efd642d70e6bea" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4b/03/c75c6ad46be41c16f4cfe0352a2d1450546f3c09ad2c9d341110cd87b025/orjson-3.10.18-cp313-cp313-win_amd64.whl", hash = "sha256:aed411bcb68bf62e85588f2a7e03a6082cc42e5a2796e06e72a962d7c6310b52" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c2/28/f53038a5a72cc4fd0b56c1eafb4ef64aec9685460d5ac34de98ca78b6e29/orjson-3.10.18-cp313-cp313-win_arm64.whl", hash = "sha256:f54c1385a0e6aba2f15a40d703b858bedad36ded0491e55d35d905b2c34a4cc3" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "ormsgpack"
feat: bump Python minimum from 3.12 to 3.13, drop strenum backport (#14767) Closes #14753 ## What changed | File | Change | |---|---| | `pyproject.toml` | `requires-python` → `>=3.13,<3.15`; remove `strenum==0.4.15` | | `Dockerfile` | `uv python install 3.13`, `uv sync --python 3.13` | | `.github/workflows/tests.yml` | `uv sync --python 3.13` on both matrix legs | | `CLAUDE.md` | dev setup command + requirements note updated | | `deepdoc/parser/mineru_parser.py` | `from strenum import StrEnum` → `from enum import StrEnum` | | `agent/tools/code_exec.py` | same | `StrEnum` has been in the stdlib since Python 3.11 — the `strenum` backport package is no longer needed once the floor is 3.13. ## Why uv.lock is not regenerated `uv lock --python 3.13` fails because: 1. The infiniflow/graspologic fork pins `numpy>=1.26.4,<2.0.0` 2. `tensorflow-cpu>=2.20.0` (the first release with cp313 wheels) depends on `ml-dtypes>=0.5.1`, which requires `numpy>=2.1.0` 3. These two constraints are irreconcilable on Python 3.13 The lockfile regeneration requires loosening the `numpy` upper bound in the `infiniflow/graspologic` fork. Once that fork commit is updated and the SHA in `pyproject.toml:49` is bumped, `uv lock --python 3.13` will succeed. ## RFC corrections Two claims in the original RFC (#14753) did not hold up under code review: - **"graspologic hard-blocks 3.13"** — the infiniflow fork at the pinned commit has no `<3.13` Python constraint. The blocker is the transitive `numpy<2.0.0` conflict with tensorflow-cpu's test dependency, not a direct Python version cap. - **"free-threading throughput gains for I/O-bound workload"** — Python 3.13 free-threading requires a special `--disable-gil` build and provides no benefit for async I/O code (the GIL is already released during I/O). The real motivation is forward compatibility and improved error messages.
2026-05-15 08:40:53 +02:00
version = "1.12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/12/0c/f1761e21486942ab9bb6feaebc610fa074f7c5e496e6962dea5873348077/ormsgpack-1.12.2.tar.gz", hash = "sha256:944a2233640273bee67521795a73cf1e959538e0dfb7ac635505010455e53b33" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/eb/29/bb0eba3288c0449efbb013e9c6f58aea79cf5cb9ee1921f8865f04c1a9d7/ormsgpack-1.12.2-cp313-cp313-macosx_10_12_x86_64.macosx_11_0_arm64.macosx_10_12_universal2.whl", hash = "sha256:5ea60cb5f210b1cfbad8c002948d73447508e629ec375acb82910e3efa8ff355" },
{ url = "https://mirrors.aliyun.com/pypi/packages/6e/31/5efa31346affdac489acade2926989e019e8ca98129658a183e3add7af5e/ormsgpack-1.12.2-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f3601f19afdbea273ed70b06495e5794606a8b690a568d6c996a90d7255e51c1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/eb/56/d0087278beef833187e0167f8527235ebe6f6ffc2a143e9de12a98b1ce87/ormsgpack-1.12.2-cp313-cp313-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:29a9f17a3dac6054c0dce7925e0f4995c727f7c41859adf9b5572180f640d172" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1c/a2/072343e1413d9443e5a252a8eb591c2d5b1bffbe5e7bfc78c069361b92eb/ormsgpack-1.12.2-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:39c1bd2092880e413902910388be8715f70b9f15f20779d44e673033a6146f2d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a2/8b/a0da3b98a91d41187a63b02dda14267eefc2a74fcb43cc2701066cf1510e/ormsgpack-1.12.2-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:50b7249244382209877deedeee838aef1542f3d0fc28b8fe71ca9d7e1896a0d7" },
{ url = "https://mirrors.aliyun.com/pypi/packages/19/bb/6d226bc4cf9fc20d8eb1d976d027a3f7c3491e8f08289a2e76abe96a65f3/ormsgpack-1.12.2-cp313-cp313-musllinux_1_2_armv7l.whl", hash = "sha256:5af04800d844451cf102a59c74a841324868d3f1625c296a06cc655c542a6685" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fb/f1/bb2c7223398543dedb3dbf8bb93aaa737b387de61c5feaad6f908841b782/ormsgpack-1.12.2-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:cec70477d4371cd524534cd16472d8b9cc187e0e3043a8790545a9a9b296c258" },
{ url = "https://mirrors.aliyun.com/pypi/packages/7b/e8/0fb45f57a2ada1fed374f7494c8cd55e2f88ccd0ab0a669aa3468716bf5f/ormsgpack-1.12.2-cp313-cp313-win_amd64.whl", hash = "sha256:21f4276caca5c03a818041d637e4019bc84f9d6ca8baa5ea03e5cc8bf56140e9" },
{ url = "https://mirrors.aliyun.com/pypi/packages/7a/d4/0cfeea1e960d550a131001a7f38a5132c7ae3ebde4c82af1f364ccc5d904/ormsgpack-1.12.2-cp313-cp313-win_arm64.whl", hash = "sha256:baca4b6773d20a82e36d6fd25f341064244f9f86a13dead95dd7d7f996f51709" },
]
[[package]]
name = "oss2"
version = "2.19.1"
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
dependencies = [
{ name = "aliyun-python-sdk-core" },
{ name = "aliyun-python-sdk-kms" },
{ name = "crcmod" },
{ name = "pycryptodome" },
{ name = "requests" },
{ name = "six" },
Security: do not use litellm 1.82.7 and 1.82.8 (#13768) ### What problem does this PR solve? See [issue](https://github.com/BerriAI/litellm/issues/24518) from Litellm. Upgraded from `1.81.15` to `1.82.6`, so RAGFlow is safe as always. ### Type of change - [x] Security Co-authored-by: Jin Hai <haijin.chn@gmail.com> Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
2026-03-25 22:39:33 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/df/b5/f2cb1950dda46ac2284d6c950489fdacd0e743c2d79a347924d3cc44b86f/oss2-2.19.1.tar.gz", hash = "sha256:a8ab9ee7eb99e88a7e1382edc6ea641d219d585a7e074e3776e9dec9473e59c1" }
Security: do not use litellm 1.82.7 and 1.82.8 (#13768) ### What problem does this PR solve? See [issue](https://github.com/BerriAI/litellm/issues/24518) from Litellm. Upgraded from `1.81.15` to `1.82.6`, so RAGFlow is safe as always. ### Type of change - [x] Security Co-authored-by: Jin Hai <haijin.chn@gmail.com> Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
2026-03-25 22:39:33 +08:00
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "outcome"
version = "1.3.0.post0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "attrs" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/98/df/77698abfac98571e65ffeb0c1fba8ffd692ab8458d617a0eed7d9a8d38f2/outcome-1.3.0.post0.tar.gz", hash = "sha256:9dcf02e65f2971b80047b377468e72a268e15c0af3cf1238e6ff14f7f91143b8" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/55/8b/5ab7257531a5d830fc8000c476e63c935488d74609b50f9384a643ec0a62/outcome-1.3.0.post0-py2.py3-none-any.whl", hash = "sha256:e771c5ce06d1415e356078d3bdd68523f284b4ce5419828922b6871e65eda82b" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "packaging"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "25.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a1/d4/1fc4078c65507b51b96ca8f8c3ba19e6a61c8253c72794544580a7b6c24d/packaging-25.0.tar.gz", hash = "sha256:d443872c98d677bf60f6a1f2f8c1cb748e8fe762d2bf9d3148b5599295b0fc4f" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/20/12/38679034af332785aac8774540895e234f4d07f7545804097de4b666afd8/packaging-25.0-py3-none-any.whl", hash = "sha256:29572ef2b1f17581046b3a2227d5c611fb25ec70ca1ba8554b24b0e69331a484" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "pandas"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "2.3.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "numpy" },
{ name = "python-dateutil" },
{ name = "pytz" },
{ name = "tzdata" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/33/01/d40b85317f86cf08d853a4f495195c73815fdf205eef3993821720274518/pandas-2.3.3.tar.gz", hash = "sha256:e05e1af93b977f7eafa636d043f9f94c7ee3ac81af99c13508215942e64c993b" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/cd/4b/18b035ee18f97c1040d94debd8f2e737000ad70ccc8f5513f4eefad75f4b/pandas-2.3.3-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:56851a737e3470de7fa88e6131f41281ed440d29a9268dcbf0002da5ac366713" },
{ url = "https://mirrors.aliyun.com/pypi/packages/31/94/72fac03573102779920099bcac1c3b05975c2cb5f01eac609faf34bed1ca/pandas-2.3.3-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:bdcd9d1167f4885211e401b3036c0c8d9e274eee67ea8d0758a256d60704cfe8" },
{ url = "https://mirrors.aliyun.com/pypi/packages/16/87/9472cf4a487d848476865321de18cc8c920b8cab98453ab79dbbc98db63a/pandas-2.3.3-cp313-cp313-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:e32e7cc9af0f1cc15548288a51a3b681cc2a219faa838e995f7dc53dbab1062d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/15/07/284f757f63f8a8d69ed4472bfd85122bd086e637bf4ed09de572d575a693/pandas-2.3.3-cp313-cp313-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:318d77e0e42a628c04dc56bcef4b40de67918f7041c2b061af1da41dcff670ac" },
{ url = "https://mirrors.aliyun.com/pypi/packages/33/81/a3afc88fca4aa925804a27d2676d22dcd2031c2ebe08aabd0ae55b9ff282/pandas-2.3.3-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:4e0a175408804d566144e170d0476b15d78458795bb18f1304fb94160cabf40c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8d/0f/b4d4ae743a83742f1153464cf1a8ecfafc3ac59722a0b5c8602310cb7158/pandas-2.3.3-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:93c2d9ab0fc11822b5eece72ec9587e172f63cff87c00b062f6e37448ced4493" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4f/c7/e54682c96a895d0c808453269e0b5928a07a127a15704fedb643e9b0a4c8/pandas-2.3.3-cp313-cp313-win_amd64.whl", hash = "sha256:f8bfc0e12dc78f777f323f55c58649591b2cd0c43534e8355c51d3fede5f4dee" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f9/ca/3f8d4f49740799189e1395812f3bf23b5e8fc7c190827d55a610da72ce55/pandas-2.3.3-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:75ea25f9529fdec2d2e93a42c523962261e567d250b0013b16210e1d40d7c2e5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0e/5a/f43efec3e8c0cc92c4663ccad372dbdff72b60bdb56b2749f04aa1d07d7e/pandas-2.3.3-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:74ecdf1d301e812db96a465a525952f4dde225fdb6d8e5a521d47e1f42041e21" },
{ url = "https://mirrors.aliyun.com/pypi/packages/46/b1/85331edfc591208c9d1a63a06baa67b21d332e63b7a591a5ba42a10bb507/pandas-2.3.3-cp313-cp313t-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:6435cb949cb34ec11cc9860246ccb2fdc9ecd742c12d3304989017d53f039a78" },
{ url = "https://mirrors.aliyun.com/pypi/packages/44/23/78d645adc35d94d1ac4f2a3c4112ab6f5b8999f4898b8cdf01252f8df4a9/pandas-2.3.3-cp313-cp313t-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:900f47d8f20860de523a1ac881c4c36d65efcb2eb850e6948140fa781736e110" },
{ url = "https://mirrors.aliyun.com/pypi/packages/53/da/d10013df5e6aaef6b425aa0c32e1fc1f3e431e4bcabd420517dceadce354/pandas-2.3.3-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:a45c765238e2ed7d7c608fc5bc4a6f88b642f2f01e70c0c23d2224dd21829d86" },
{ url = "https://mirrors.aliyun.com/pypi/packages/bd/17/e756653095a083d8a37cbd816cb87148debcfcd920129b25f99dd8d04271/pandas-2.3.3-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:c4fc4c21971a1a9f4bdb4c73978c7f7256caa3e62b323f70d6cb80db583350bc" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "parameterized"
version = "0.9.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ea/49/00c0c0cc24ff4266025a53e41336b79adaa5a4ebfad214f433d623f9865e/parameterized-0.9.0.tar.gz", hash = "sha256:7fc905272cefa4f364c1a3429cbbe9c0f98b793988efb5bf90aac80f08db09b1" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/00/2f/804f58f0b856ab3bf21617cccf5b39206e6c4c94c2cd227bde125ea6105f/parameterized-0.9.0-py2.py3-none-any.whl", hash = "sha256:4e0758e3d41bea3bbd05ec14fc2c24736723f243b28d702081aef438c9372b1b" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: add local & ssh provider in admin panel (#15039) ### What problem does this PR solve? Feat: add local & ssh provider in admin panel ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-20 16:56:20 +08:00
[[package]]
name = "paramiko"
version = "5.0.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: add local & ssh provider in admin panel (#15039) ### What problem does this PR solve? Feat: add local & ssh provider in admin panel ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-20 16:56:20 +08:00
dependencies = [
{ name = "bcrypt" },
{ name = "cryptography" },
{ name = "invoke" },
{ name = "pynacl" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/62/93/dcc25d52f49022ae6175d15e6bd751f1acc99b98bc61fc55e5155a7be2e7/paramiko-5.0.0.tar.gz", hash = "sha256:36763b5b95c2a0dcfdf1abc48e48156ee425b21efe2f0e787c2dd5a95c0e5e79" }
Feat: add local & ssh provider in admin panel (#15039) ### What problem does this PR solve? Feat: add local & ssh provider in admin panel ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-20 16:56:20 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/82/5b/eadf6d45de38d30ab603f49393b6cd2cbe7e233af8cf90197e32782b68a9/paramiko-5.0.0-py3-none-any.whl", hash = "sha256:b7044611c30140d9a75261653210e2002977b71a0497ff3ba0d98d7edbf62f7c" },
Feat: add local & ssh provider in admin panel (#15039) ### What problem does this PR solve? Feat: add local & ssh provider in admin panel ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-20 16:56:20 +08:00
]
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
[[package]]
name = "patchright"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.60.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
dependencies = [
{ name = "greenlet" },
{ name = "pyee" },
]
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/71/26/c1e858fd1acc63e410b3d33243955f36d2a0814487b97a7aa604ad2baffd/patchright-1.60.1-py3-none-macosx_10_13_x86_64.whl", hash = "sha256:e9492100d4e2a85ff92fc3a668dd16dee03f21df6e559c7b9f7c71e86ff48c6b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/55/dd/2dd8e4e02489ec8fd57ad93dec9ef444b6f42adcc4fe95df30237c92841d/patchright-1.60.1-py3-none-macosx_11_0_arm64.whl", hash = "sha256:20bd806df2469b451ccd2ea10f5f944ceb0e0d83c716f5752b0c956c1ee59476" },
{ url = "https://mirrors.aliyun.com/pypi/packages/54/cc/0fa0bedec61045fd9068682e3695557b622c483f829d341093879ec8dbd9/patchright-1.60.1-py3-none-macosx_11_0_universal2.whl", hash = "sha256:9fd15a64c0ca80740dc2a3f41cda336a06a2ed6068d0ab893172654290b06e6b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ce/c2/4b8f69de0a20d90792980c43c0e60b10b801e08cf0224ccc8a8266e1fffb/patchright-1.60.1-py3-none-manylinux1_x86_64.whl", hash = "sha256:547e7bfb813102309789cc42933780e5fdf7c4727de59fb2791e64bd1298a7f3" },
{ url = "https://mirrors.aliyun.com/pypi/packages/db/fc/9fd6a70818cf0bc3b62574af483d762963cb65ca0a369826a0536faffe41/patchright-1.60.1-py3-none-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:023945a2fd30219a284721ca36385bd44075ae7b53071dc1da38036b6dbe88ec" },
{ url = "https://mirrors.aliyun.com/pypi/packages/08/bc/81fb621e5ab4131e6f324c9ba6cb2a9f3b146c92f12484bec95efe8c8347/patchright-1.60.1-py3-none-win32.whl", hash = "sha256:05b98a6afdbe7e6645fe223009c47cc8e7859df55fd8ce9d8a9925b3389b0ee1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/74/8e/fff80350ed2c2c1f62145d667070799c60ca9e9b28d54e4f751f0b4f8da6/patchright-1.60.1-py3-none-win_amd64.whl", hash = "sha256:51b306ed55cd58f1bca24641458f5c9f7e86a1f1727dcffdead669cfe4c0a485" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ee/b8/b6d1bfe98a420c1ccfb2f23a7bb2b4cdb40170907bae638e7ae92abd287c/patchright-1.60.1-py3-none-win_arm64.whl", hash = "sha256:f795728c1e27fc226dbe203c1aec713a537f01be963474fe0f3691f5e6457f9f" },
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "patsy"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "1.0.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "numpy" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/be/44/ed13eccdd0519eff265f44b670d46fbb0ec813e2274932dc1c0e48520f7d/patsy-1.0.2.tar.gz", hash = "sha256:cdc995455f6233e90e22de72c37fcadb344e7586fb83f06696f54d92f8ce74c0" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/f1/70/ba4b949bdc0490ab78d545459acd7702b211dfccf7eb89bbc1060f52818d/patsy-1.0.2-py2.py3-none-any.whl", hash = "sha256:37bfddbc58fcf0362febb5f54f10743f8b21dd2aa73dec7e7ef59d1b02ae668a" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
[[package]]
name = "pcodedmp"
version = "1.2.6"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
dependencies = [
{ name = "oletools" },
{ name = "win-unicode-console", marker = "platform_python_implementation != 'PyPy' and sys_platform == 'win32'" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/3d/20/6d461e29135f474408d0d7f95b2456a9ba245560768ee51b788af10f7429/pcodedmp-1.2.6.tar.gz", hash = "sha256:025f8c809a126f45a082ffa820893e6a8d990d9d7ddb68694b5a9f0a6dbcd955" }
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ba/72/b380fb5c89d89c3afafac8cf02a71a45f4f4a4f35531ca949a34683962d1/pcodedmp-1.2.6-py2.py3-none-any.whl", hash = "sha256:4441f7c0ab4cbda27bd4668db3b14f36261d86e5059ce06c0828602cbe1c4278" },
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "pdfminer-six"
version = "20221105"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "charset-normalizer" },
{ name = "cryptography" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ac/6e/89c532d108e362cbaf76fdb972e7a5e85723c225f08e1646fb86878d4f7f/pdfminer.six-20221105.tar.gz", hash = "sha256:8448ab7b939d18b64820478ecac5394f482d7a79f5f7eaa7703c6c959c175e1d" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/46/68/b3fb5f073bcd3df9143a3520289c147351bfa3c1b096d44081f38fd1c247/pdfminer.six-20221105-py3-none-any.whl", hash = "sha256:1eaddd712d5b2732f8ac8486824533514f8ba12a0787b3d5fe1e686cd826532d" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "pdfplumber"
version = "0.10.4"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "pdfminer-six" },
{ name = "pillow" },
{ name = "pypdfium2" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/25/b2/a3ebd1987165b088dfa328fc1ddbf621b62f1785a4daafb4090c91246b61/pdfplumber-0.10.4.tar.gz", hash = "sha256:1c83a4e1fe75525ce1f161fa55a8142209a2da69b45542ce2c45be879e804fd6" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/53/5c/4523bfce8ba473b0e33931f9638f69c3573b3b72b0c63c73d779848d182f/pdfplumber-0.10.4-py3-none-any.whl", hash = "sha256:c8f200259703324cd39a5c93b181a0d2370a6b2b6da670c117e75c3da6aca4a4" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "peewee"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "3.19.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/88/b0/79462b42e89764998756e0557f2b58a15610a5b4512fbbcccae58fba7237/peewee-3.19.0.tar.gz", hash = "sha256:f88292a6f0d7b906cb26bca9c8599b8f4d8920ebd36124400d0cbaaaf915511f" }
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/1a/41/19c65578ef9a54b3083253c68a607f099642747168fe00f3a2bceb7c3a34/peewee-3.19.0-py3-none-any.whl", hash = "sha256:de220b94766e6008c466e00ce4ba5299b9a832117d9eb36d45d0062f3cfd7417" },
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
[[package]]
name = "pfzy"
version = "0.3.4"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/d9/5a/32b50c077c86bfccc7bed4881c5a2b823518f5450a30e639db5d3711952e/pfzy-0.3.4.tar.gz", hash = "sha256:717ea765dd10b63618e7298b2d98efd819e0b30cd5905c9707223dceeb94b3f1" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/8c/d7/8ff98376b1acc4503253b685ea09981697385ce344d4e3935c2af49e044d/pfzy-0.3.4-py3-none-any.whl", hash = "sha256:5f50d5b2b3207fa72e7ec0ef08372ef652685470974a107d0d4999fc5a903a96" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "pillow"
Chore(deps-dev): Bump pillow from 12.1.1 to 12.2.0 (#14578) As title
2026-05-06 11:08:38 +08:00
version = "12.2.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/8c/21/c2bcdd5906101a30244eaffc1b6e6ce71a31bd0742a01eb89e660ebfac2d/pillow-12.2.0.tar.gz", hash = "sha256:a830b1a40919539d07806aa58e1b114df53ddd43213d9c8b75847eee6c0182b5" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/4a/01/53d10cf0dbad820a8db274d259a37ba50b88b24768ddccec07355382d5ad/pillow-12.2.0-cp313-cp313-ios_13_0_arm64_iphoneos.whl", hash = "sha256:8297651f5b5679c19968abefd6bb84d95fe30ef712eb1b2d9b2d31ca61267f4c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0f/98/f3a6657ecb698c937f6c76ee564882945f29b79bad496abcba0e84659ec5/pillow-12.2.0-cp313-cp313-ios_13_0_arm64_iphonesimulator.whl", hash = "sha256:50d8520da2a6ce0af445fa6d648c4273c3eeefbc32d7ce049f22e8b5c3daecc2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/69/bc/8986948f05e3ea490b8442ea1c1d4d990b24a7e43d8a51b2c7d8b1dced36/pillow-12.2.0-cp313-cp313-ios_13_0_x86_64_iphonesimulator.whl", hash = "sha256:766cef22385fa1091258ad7e6216792b156dc16d8d3fa607e7545b2b72061f1c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/34/46/6c717baadcd62bc8ed51d238d521ab651eaa74838291bda1f86fe1f864c9/pillow-12.2.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:5d2fd0fa6b5d9d1de415060363433f28da8b1526c1c129020435e186794b3795" },
{ url = "https://mirrors.aliyun.com/pypi/packages/71/43/905a14a8b17fdb1ccb58d282454490662d2cb89a6bfec26af6d3520da5ec/pillow-12.2.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:56b25336f502b6ed02e889f4ece894a72612fe885889a6e8c4c80239ff6e5f5f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/73/dd/42107efcb777b16fa0393317eac58f5b5cf30e8392e266e76e51cff28c3d/pillow-12.2.0-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:f1c943e96e85df3d3478f7b691f229887e143f81fedab9b20205349ab04d73ed" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a8/68/b93e09e5e8549019e61acf49f65b1a8530765a7f812c77a7461bca7e4494/pillow-12.2.0-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:03f6fab9219220f041c74aeaa2939ff0062bd5c364ba9ce037197f4c6d498cd9" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4b/6e/3ccb54ce8ec4ddd1accd2d89004308b7b0b21c4ac3d20fa70af4760a4330/pillow-12.2.0-cp313-cp313-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:5cdfebd752ec52bf5bb4e35d9c64b40826bc5b40a13df7c3cda20a2c03a0f5ed" },
{ url = "https://mirrors.aliyun.com/pypi/packages/67/ee/21d4e8536afd1a328f01b359b4d3997b291ffd35a237c877b331c1c3b71c/pillow-12.2.0-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:eedf4b74eda2b5a4b2b2fb4c006d6295df3bf29e459e198c90ea48e130dc75c3" },
{ url = "https://mirrors.aliyun.com/pypi/packages/78/5f/e9f86ab0146464e8c133fe85df987ed9e77e08b29d8d35f9f9f4d6f917ba/pillow-12.2.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:00a2865911330191c0b818c59103b58a5e697cae67042366970a6b6f1b20b7f9" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ed/1e/409007f56a2fdce61584fd3acbc2bbc259857d555196cedcadc68c015c82/pillow-12.2.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:1e1757442ed87f4912397c6d35a0db6a7b52592156014706f17658ff58bbf795" },
{ url = "https://mirrors.aliyun.com/pypi/packages/23/c4/7349421080b12fb35414607b8871e9534546c128a11965fd4a7002ccfbee/pillow-12.2.0-cp313-cp313-win32.whl", hash = "sha256:144748b3af2d1b358d41286056d0003f47cb339b8c43a9ea42f5fea4d8c66b6e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3f/82/8a3739a5e470b3c6cbb1d21d315800d8e16bff503d1f16b03a4ec3212786/pillow-12.2.0-cp313-cp313-win_amd64.whl", hash = "sha256:390ede346628ccc626e5730107cde16c42d3836b89662a115a921f28440e6a3b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c3/25/f968f618a062574294592f668218f8af564830ccebdd1fa6200f598e65c5/pillow-12.2.0-cp313-cp313-win_arm64.whl", hash = "sha256:8023abc91fba39036dbce14a7d6535632f99c0b857807cbbbf21ecc9f4717f06" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4d/a4/b342930964e3cb4dce5038ae34b0eab4653334995336cd486c5a8c25a00c/pillow-12.2.0-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:042db20a421b9bafecc4b84a8b6e444686bd9d836c7fd24542db3e7df7baad9b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9f/de/23198e0a65a9cf06123f5435a5d95cea62a635697f8f03d134d3f3a96151/pillow-12.2.0-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:dd025009355c926a84a612fecf58bb315a3f6814b17ead51a8e48d3823d9087f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/01/a6/1265e977f17d93ea37aa28aa81bad4fa597933879fac2520d24e021c8da3/pillow-12.2.0-cp313-cp313t-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:88ddbc66737e277852913bd1e07c150cc7bb124539f94c4e2df5344494e0a612" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3c/83/5982eb4a285967baa70340320be9f88e57665a387e3a53a7f0db8231a0cd/pillow-12.2.0-cp313-cp313t-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:d362d1878f00c142b7e1a16e6e5e780f02be8195123f164edf7eddd911eefe7c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4e/48/6ffc514adce69f6050d0753b1a18fd920fce8cac87620d5a31231b04bfc5/pillow-12.2.0-cp313-cp313t-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:2c727a6d53cb0018aadd8018c2b938376af27914a68a492f59dfcaca650d5eea" },
{ url = "https://mirrors.aliyun.com/pypi/packages/36/a3/f9a77144231fb8d40ee27107b4463e205fa4677e2ca2548e14da5cf18dce/pillow-12.2.0-cp313-cp313t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:efd8c21c98c5cc60653bcb311bef2ce0401642b7ce9d09e03a7da87c878289d4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c1/fc/ac4ee3041e7d5a565e1c4fd72a113f03b6394cc72ab7089d27608f8aaccb/pillow-12.2.0-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:9f08483a632889536b8139663db60f6724bfcb443c96f1b18855860d7d5c0fd4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c0/a8/27fb307055087f3668f6d0a8ccb636e7431d56ed0750e07a60547b1e083e/pillow-12.2.0-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:dac8d77255a37e81a2efcbd1fc05f1c15ee82200e6c240d7e127e25e365c39ea" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ad/4b/926ab182c07fccae9fcb120043464e1ff1564775ec8864f21a0ebce6ac25/pillow-12.2.0-cp313-cp313t-win32.whl", hash = "sha256:ee3120ae9dff32f121610bb08e4313be87e03efeadfc6c0d18f89127e24d0c24" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c2/c4/f9e476451a098181b30050cc4c9a3556b64c02cf6497ea421ac047e89e4b/pillow-12.2.0-cp313-cp313t-win_amd64.whl", hash = "sha256:325ca0528c6788d2a6c3d40e3568639398137346c3d6e66bb61db96b96511c98" },
{ url = "https://mirrors.aliyun.com/pypi/packages/00/a4/285f12aeacbe2d6dc36c407dfbbe9e96d4a80b0fb710a337f6d2ad978c75/pillow-12.2.0-cp313-cp313t-win_arm64.whl", hash = "sha256:2e5a76d03a6c6dcef67edabda7a52494afa4035021a79c8558e14af25313d453" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "platformdirs"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "4.5.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/cf/86/0248f086a84f01b37aaec0fa567b397df1a119f73c16f6c7a9aac73ea309/platformdirs-4.5.1.tar.gz", hash = "sha256:61d5cdcc6065745cdd94f0f878977f8de9437be93de97c1c12f853c9c0cdcbda" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/cb/28/3bfe2fa5a7b9c46fe7e13c97bda14c895fb10fa2ebf1d0abb90e0cea7ee1/platformdirs-4.5.1-py3-none-any.whl", hash = "sha256:d03afa3963c806a9bed9d5125c8f4cb2fdaf74a55ab60e5d59b3fde758104d31" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "playwright"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.57.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "greenlet" },
{ name = "pyee" },
]
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ed/b6/e17543cea8290ae4dced10be21d5a43c360096aa2cce0aa7039e60c50df3/playwright-1.57.0-py3-none-macosx_10_13_x86_64.whl", hash = "sha256:9351c1ac3dfd9b3820fe7fc4340d96c0d3736bb68097b9b7a69bd45d25e9370c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8b/04/ef95b67e1ff59c080b2effd1a9a96984d6953f667c91dfe9d77c838fc956/playwright-1.57.0-py3-none-macosx_11_0_arm64.whl", hash = "sha256:a4a9d65027bce48eeba842408bcc1421502dfd7e41e28d207e94260fa93ca67e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/60/bd/5563850322a663956c927eefcf1457d12917e8f118c214410e815f2147d1/playwright-1.57.0-py3-none-macosx_11_0_universal2.whl", hash = "sha256:99104771abc4eafee48f47dac2369e0015516dc1ce8c409807d2dd440828b9a4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/56/61/3a803cb5ae0321715bfd5247ea871d25b32c8f372aeb70550a90c5f586df/playwright-1.57.0-py3-none-manylinux1_x86_64.whl", hash = "sha256:284ed5a706b7c389a06caa431b2f0ba9ac4130113c3a779767dda758c2497bb1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/83/d7/b72eb59dfbea0013a7f9731878df8c670f5f35318cedb010c8a30292c118/playwright-1.57.0-py3-none-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:38a1bae6c0a07839cdeaddbc0756b3b2b85e476c07945f64ece08f1f956a86f1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e4/09/3fc9ebd7c95ee54ba6a68d5c0bc23e449f7235f4603fc60534a364934c16/playwright-1.57.0-py3-none-win32.whl", hash = "sha256:1dd93b265688da46e91ecb0606d36f777f8eadcf7fbef12f6426b20bf0c9137c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/58/d4/dcdfd2a33096aeda6ca0d15584800443dd2be64becca8f315634044b135b/playwright-1.57.0-py3-none-win_amd64.whl", hash = "sha256:6caefb08ed2c6f29d33b8088d05d09376946e49a73be19271c8cd5384b82b14c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/6a/60/fe31d7e6b8907789dcb0584f88be741ba388413e4fbce35f1eba4e3073de/playwright-1.57.0-py3-none-win_arm64.whl", hash = "sha256:5f065f5a133dbc15e6e7c71e7bc04f258195755b1c32a432b792e28338c8335e" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
fix: upgrade crawl4ai to 0.8.0 (CVE-2026-26217) (#15415) ## Summary Upgrade crawl4ai from 0.7.6 to 0.8.0 to fix CVE-2026-26217. ## Vulnerability | Field | Value | |-------|-------| | **ID** | CVE-2026-26217 | | **Severity** | CRITICAL | | **Scanner** | trivy | | **Rule** | `CVE-2026-26217` | | **File** | `uv.lock` | | **Assessment** | Likely exploitable | **Description**: Crawl4AI Has Local File Inclusion in Docker API via file:// URLs ## Evidence **Scanner confirmation**: trivy rule `CVE-2026-26217` flagged this pattern. **Production code**: This file is in the production codebase, not test-only code. ## Threat Model Context This is a web service - vulnerabilities in request handlers are directly exploitable by remote attackers. ## Changes - `pyproject.toml` - `uv.lock` ## Verification - [x] Build passes - [x] Scanner re-scan confirms fix - [x] LLM code review passed --- *This change addresses a pattern flagged by static analysis. The code path handles user-influenced input and the fix reduces the attack surface against both manual and automated exploitation.* --- *Automated security fix by [OrbisAI Security](https://orbisappsec.com)*
2026-05-29 19:08:41 +05:30
[[package]]
name = "playwright-stealth"
version = "2.0.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
fix: upgrade crawl4ai to 0.8.0 (CVE-2026-26217) (#15415) ## Summary Upgrade crawl4ai from 0.7.6 to 0.8.0 to fix CVE-2026-26217. ## Vulnerability | Field | Value | |-------|-------| | **ID** | CVE-2026-26217 | | **Severity** | CRITICAL | | **Scanner** | trivy | | **Rule** | `CVE-2026-26217` | | **File** | `uv.lock` | | **Assessment** | Likely exploitable | **Description**: Crawl4AI Has Local File Inclusion in Docker API via file:// URLs ## Evidence **Scanner confirmation**: trivy rule `CVE-2026-26217` flagged this pattern. **Production code**: This file is in the production codebase, not test-only code. ## Threat Model Context This is a web service - vulnerabilities in request handlers are directly exploitable by remote attackers. ## Changes - `pyproject.toml` - `uv.lock` ## Verification - [x] Build passes - [x] Scanner re-scan confirms fix - [x] LLM code review passed --- *This change addresses a pattern flagged by static analysis. The code path handles user-influenced input and the fix reduces the attack surface against both manual and automated exploitation.* --- *Automated security fix by [OrbisAI Security](https://orbisappsec.com)*
2026-05-29 19:08:41 +05:30
dependencies = [
{ name = "playwright" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a4/db/6ade5d539c7d151b9defc78fafa8b65aa52352617d0e7699b47008bd801f/playwright_stealth-2.0.3.tar.gz", hash = "sha256:1d8e488fbdd8f190f1269ea8cf5d57d14df3a9f1af1001c41ee3588b2aac3133" }
fix: upgrade crawl4ai to 0.8.0 (CVE-2026-26217) (#15415) ## Summary Upgrade crawl4ai from 0.7.6 to 0.8.0 to fix CVE-2026-26217. ## Vulnerability | Field | Value | |-------|-------| | **ID** | CVE-2026-26217 | | **Severity** | CRITICAL | | **Scanner** | trivy | | **Rule** | `CVE-2026-26217` | | **File** | `uv.lock` | | **Assessment** | Likely exploitable | **Description**: Crawl4AI Has Local File Inclusion in Docker API via file:// URLs ## Evidence **Scanner confirmation**: trivy rule `CVE-2026-26217` flagged this pattern. **Production code**: This file is in the production codebase, not test-only code. ## Threat Model Context This is a web service - vulnerabilities in request handlers are directly exploitable by remote attackers. ## Changes - `pyproject.toml` - `uv.lock` ## Verification - [x] Build passes - [x] Scanner re-scan confirms fix - [x] LLM code review passed --- *This change addresses a pattern flagged by static analysis. The code path handles user-influenced input and the fix reduces the attack surface against both manual and automated exploitation.* --- *Automated security fix by [OrbisAI Security](https://orbisappsec.com)*
2026-05-29 19:08:41 +05:30
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/b5/10/607c409712c02a26c4cb794820514cb7fdaaeac15fb05bed917fb8a354b3/playwright_stealth-2.0.3-py3-none-any.whl", hash = "sha256:1887ade423ab7ff8ae16d363a30a38de0b5817e1e4a29d47b74bf3a0e3dbfcb4" },
fix: upgrade crawl4ai to 0.8.0 (CVE-2026-26217) (#15415) ## Summary Upgrade crawl4ai from 0.7.6 to 0.8.0 to fix CVE-2026-26217. ## Vulnerability | Field | Value | |-------|-------| | **ID** | CVE-2026-26217 | | **Severity** | CRITICAL | | **Scanner** | trivy | | **Rule** | `CVE-2026-26217` | | **File** | `uv.lock` | | **Assessment** | Likely exploitable | **Description**: Crawl4AI Has Local File Inclusion in Docker API via file:// URLs ## Evidence **Scanner confirmation**: trivy rule `CVE-2026-26217` flagged this pattern. **Production code**: This file is in the production codebase, not test-only code. ## Threat Model Context This is a web service - vulnerabilities in request handlers are directly exploitable by remote attackers. ## Changes - `pyproject.toml` - `uv.lock` ## Verification - [x] Build passes - [x] Scanner re-scan confirms fix - [x] LLM code review passed --- *This change addresses a pattern flagged by static analysis. The code path handles user-influenced input and the fix reduces the attack surface against both manual and automated exploitation.* --- *Automated security fix by [OrbisAI Security](https://orbisappsec.com)*
2026-05-29 19:08:41 +05:30
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "pluggy"
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
version = "1.6.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f9/e2/3e91f31a7d2b083fe6ef3fa267035b518369d9511ffab804f839851d2779/pluggy-1.6.0.tar.gz", hash = "sha256:7dcc130b76258d33b90f61b658791dede3486c3e6bfb003ee5c9bfb396dd22f3" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/54/20/4d324d65cc6d9205fabedc306948156824eb9f0ee1633355a8f7ec5c66bf/pluggy-1.6.0-py3-none-any.whl", hash = "sha256:e920276dd6813095e9377c0bc5566d94c932c33b27a3e3945d8389c374dd4746" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: Support tool calling in Generate component (#7572) ### What problem does this PR solve? Hello, our use case requires LLM agent to invoke some tools, so I made a simple implementation here. This PR does two things: 1. A simple plugin mechanism based on `pluginlib`: This mechanism lives in the `plugin` directory. It will only load plugins from `plugin/embedded_plugins` for now. A sample plugin `bad_calculator.py` is placed in `plugin/embedded_plugins/llm_tools`, it accepts two numbers `a` and `b`, then give a wrong result `a + b + 100`. In the future, it can load plugins from external location with little code change. Plugins are divided into different types. The only plugin type supported in this PR is `llm_tools`, which must implement the `LLMToolPlugin` class in the `plugin/llm_tool_plugin.py`. More plugin types can be added in the future. 2. A tool selector in the `Generate` component: Added a tool selector to select one or more tools for LLM: ![image](https://github.com/user-attachments/assets/74a21fdf-9333-4175-991b-43df6524c5dc) And with the `bad_calculator` tool, it results this with the `qwen-max` model: ![image](https://github.com/user-attachments/assets/93aff9c4-8550-414a-90a2-1a15a5249d94) ### Type of change - [ ] Bug Fix (non-breaking change which fixes an issue) - [x] New Feature (non-breaking change which adds functionality) - [ ] Documentation Update - [ ] Refactoring - [ ] Performance Improvement - [ ] Other (please describe): Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
2025-05-16 16:32:19 +08:00
[[package]]
name = "pluginlib"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "0.11.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Support tool calling in Generate component (#7572) ### What problem does this PR solve? Hello, our use case requires LLM agent to invoke some tools, so I made a simple implementation here. This PR does two things: 1. A simple plugin mechanism based on `pluginlib`: This mechanism lives in the `plugin` directory. It will only load plugins from `plugin/embedded_plugins` for now. A sample plugin `bad_calculator.py` is placed in `plugin/embedded_plugins/llm_tools`, it accepts two numbers `a` and `b`, then give a wrong result `a + b + 100`. In the future, it can load plugins from external location with little code change. Plugins are divided into different types. The only plugin type supported in this PR is `llm_tools`, which must implement the `LLMToolPlugin` class in the `plugin/llm_tool_plugin.py`. More plugin types can be added in the future. 2. A tool selector in the `Generate` component: Added a tool selector to select one or more tools for LLM: ![image](https://github.com/user-attachments/assets/74a21fdf-9333-4175-991b-43df6524c5dc) And with the `bad_calculator` tool, it results this with the `qwen-max` model: ![image](https://github.com/user-attachments/assets/93aff9c4-8550-414a-90a2-1a15a5249d94) ### Type of change - [ ] Bug Fix (non-breaking change which fixes an issue) - [x] New Feature (non-breaking change which adds functionality) - [ ] Documentation Update - [ ] Refactoring - [ ] Performance Improvement - [ ] Other (please describe): Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
2025-05-16 16:32:19 +08:00
dependencies = [
Fix: #6098 - Add validation logic for parser_config when update document (#13911) ### What problem does this PR solve? Add validation logic for parser_config. Refactor the processing flow. Before change, validation logics and update logics are mixed up - some validation logis executes followed by some update logic executes and then another such "validation-and-then-update" which is not good. After change, all validation logic executes firstly. Update logic will be executed after ALL validation logic executed. Validation logic for parameters (that come from front end) will be checked using Pydantic. For validation logic that depends on data from DB, they will be in separate methods. ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue) - [x] Refactoring
2026-04-07 11:33:05 +08:00
{ name = "packaging" },
Feat: Support tool calling in Generate component (#7572) ### What problem does this PR solve? Hello, our use case requires LLM agent to invoke some tools, so I made a simple implementation here. This PR does two things: 1. A simple plugin mechanism based on `pluginlib`: This mechanism lives in the `plugin` directory. It will only load plugins from `plugin/embedded_plugins` for now. A sample plugin `bad_calculator.py` is placed in `plugin/embedded_plugins/llm_tools`, it accepts two numbers `a` and `b`, then give a wrong result `a + b + 100`. In the future, it can load plugins from external location with little code change. Plugins are divided into different types. The only plugin type supported in this PR is `llm_tools`, which must implement the `LLMToolPlugin` class in the `plugin/llm_tool_plugin.py`. More plugin types can be added in the future. 2. A tool selector in the `Generate` component: Added a tool selector to select one or more tools for LLM: ![image](https://github.com/user-attachments/assets/74a21fdf-9333-4175-991b-43df6524c5dc) And with the `bad_calculator` tool, it results this with the `qwen-max` model: ![image](https://github.com/user-attachments/assets/93aff9c4-8550-414a-90a2-1a15a5249d94) ### Type of change - [ ] Bug Fix (non-breaking change which fixes an issue) - [x] New Feature (non-breaking change which adds functionality) - [ ] Documentation Update - [ ] Refactoring - [ ] Performance Improvement - [ ] Other (please describe): Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
2025-05-16 16:32:19 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ed/ee/d768d383cff0128ba30e7088a5d6aa8ed083e2b3ad669be6401c999bf58d/pluginlib-0.11.0.tar.gz", hash = "sha256:daece55f03d76d0674f481b0c1b73ad877229b721386a14adbe9c9991c9f3025" }
Feat: Support tool calling in Generate component (#7572) ### What problem does this PR solve? Hello, our use case requires LLM agent to invoke some tools, so I made a simple implementation here. This PR does two things: 1. A simple plugin mechanism based on `pluginlib`: This mechanism lives in the `plugin` directory. It will only load plugins from `plugin/embedded_plugins` for now. A sample plugin `bad_calculator.py` is placed in `plugin/embedded_plugins/llm_tools`, it accepts two numbers `a` and `b`, then give a wrong result `a + b + 100`. In the future, it can load plugins from external location with little code change. Plugins are divided into different types. The only plugin type supported in this PR is `llm_tools`, which must implement the `LLMToolPlugin` class in the `plugin/llm_tool_plugin.py`. More plugin types can be added in the future. 2. A tool selector in the `Generate` component: Added a tool selector to select one or more tools for LLM: ![image](https://github.com/user-attachments/assets/74a21fdf-9333-4175-991b-43df6524c5dc) And with the `bad_calculator` tool, it results this with the `qwen-max` model: ![image](https://github.com/user-attachments/assets/93aff9c4-8550-414a-90a2-1a15a5249d94) ### Type of change - [ ] Bug Fix (non-breaking change which fixes an issue) - [x] New Feature (non-breaking change which adds functionality) - [ ] Documentation Update - [ ] Refactoring - [ ] Performance Improvement - [ ] Other (please describe): Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
2025-05-16 16:32:19 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/05/a7/8d61944116824716072cdf7bc83cf262c2edc7c546cdc4c478ea72ad0dd4/pluginlib-0.11.0-py2.py3-none-any.whl", hash = "sha256:b75908a832ca053ae3112aba6424e80f406a98ebc0bf6aa3e904b1b2e85a6e01" },
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
]
[[package]]
name = "ply"
version = "3.11"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/e5/69/882ee5c9d017149285cab114ebeab373308ef0f874fcdac9beb90e0ac4da/ply-3.11.tar.gz", hash = "sha256:00c7c1aaa88358b9c765b6d3000c6eec0ba42abca5351b095321aef446081da3" }
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/a3/58/35da89ee790598a0700ea49b2a66594140f44dec458c07e8e3d4979137fc/ply-3.11-py2.py3-none-any.whl", hash = "sha256:096f9b8350b65ebd2fd1346b12452efe5b9607f7482813ffca50c22722a807ce" },
Feat: Support tool calling in Generate component (#7572) ### What problem does this PR solve? Hello, our use case requires LLM agent to invoke some tools, so I made a simple implementation here. This PR does two things: 1. A simple plugin mechanism based on `pluginlib`: This mechanism lives in the `plugin` directory. It will only load plugins from `plugin/embedded_plugins` for now. A sample plugin `bad_calculator.py` is placed in `plugin/embedded_plugins/llm_tools`, it accepts two numbers `a` and `b`, then give a wrong result `a + b + 100`. In the future, it can load plugins from external location with little code change. Plugins are divided into different types. The only plugin type supported in this PR is `llm_tools`, which must implement the `LLMToolPlugin` class in the `plugin/llm_tool_plugin.py`. More plugin types can be added in the future. 2. A tool selector in the `Generate` component: Added a tool selector to select one or more tools for LLM: ![image](https://github.com/user-attachments/assets/74a21fdf-9333-4175-991b-43df6524c5dc) And with the `bad_calculator` tool, it results this with the `qwen-max` model: ![image](https://github.com/user-attachments/assets/93aff9c4-8550-414a-90a2-1a15a5249d94) ### Type of change - [ ] Bug Fix (non-breaking change which fixes an issue) - [x] New Feature (non-breaking change which adds functionality) - [ ] Documentation Update - [ ] Refactoring - [ ] Performance Improvement - [ ] Other (please describe): Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
2025-05-16 16:32:19 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "polars-lts-cpu"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "1.33.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f4/93/a0c4200a5e0af2eee31ea79330cb1f5f4c58f604cb3de352f654e2010c81/polars_lts_cpu-1.33.1.tar.gz", hash = "sha256:0a5426d95ec9eec937a56d3e7cf7911a4b5486c42f4dbbcc9512aa706039322c" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/f1/9b/75916636b33724afabe820b0993f60dc243793421d6f680d5fcb531fe170/polars_lts_cpu-1.33.1-cp39-abi3-macosx_10_12_x86_64.whl", hash = "sha256:5db75d1b424bd8aa34c9a670a901592f1931cc94d9fb32bdd428dbaad8c33761" },
{ url = "https://mirrors.aliyun.com/pypi/packages/81/e2/dc77b81650ba0c631c06f05d8e81faacee87730600fceca372273facf77b/polars_lts_cpu-1.33.1-cp39-abi3-macosx_11_0_arm64.whl", hash = "sha256:37cf3a56cf447c69cfb3f9cd0e714d5b0c754705d7b497b9ab86cbf56e36b3e7" },
{ url = "https://mirrors.aliyun.com/pypi/packages/27/fb/4dcff801d71dfa02ec682d6b32fd0ce5339de48797f663698d5f8348ffe7/polars_lts_cpu-1.33.1-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:656b530a672fe8fbd4c212b2a8481099e5cef63e84970975619ea7c25faeb833" },
{ url = "https://mirrors.aliyun.com/pypi/packages/54/31/0474c14dce2c0507bea40069daafb848980ba7c351ad991908e51ac895fb/polars_lts_cpu-1.33.1-cp39-abi3-manylinux_2_24_aarch64.whl", hash = "sha256:64574c784380b37167b3db3a7cfdb9839cd308e89b8818859d2ffb34a9c896b2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d4/0a/5ebba9b145388ffbbd09fa84ac3cd7d336b922e34256b1417abf0a1c2fb9/polars_lts_cpu-1.33.1-cp39-abi3-win_amd64.whl", hash = "sha256:6b849e0e1485acb8ac39bf13356d280ea7c924c2b41cd548ea6e4d102d70be77" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ce/ad/bf3db68d30ac798ca31c80624709a0c03aa890e2e20e5ca987d7e55fcfc2/polars_lts_cpu-1.33.1-cp39-abi3-win_arm64.whl", hash = "sha256:c99ab56b059cee6bcabe9fb89e97f5813be1012a2251bf77f76e15c2d1cba934" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "pooch"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.8.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "packaging" },
{ name = "platformdirs" },
{ name = "requests" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/c6/77/b3d3e00c696c16cf99af81ef7b1f5fe73bd2a307abca41bd7605429fe6e5/pooch-1.8.2.tar.gz", hash = "sha256:76561f0de68a01da4df6af38e9955c4c9d1a5c90da73f7e40276a5728ec83d10" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/a8/87/77cc11c7a9ea9fd05503def69e3d18605852cd0d4b0d3b8f15bbeb3ef1d1/pooch-1.8.2-py3-none-any.whl", hash = "sha256:3529a57096f7198778a5ceefd5ac3ef0e4d06a6ddaf9fc2d609b806f25302c47" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
[[package]]
name = "portalocker"
version = "2.10.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
dependencies = [
{ name = "pywin32", marker = "sys_platform == 'win32'" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ed/d3/c6c64067759e87af98cc668c1cc75171347d0f1577fab7ca3749134e3cd4/portalocker-2.10.1.tar.gz", hash = "sha256:ef1bf844e878ab08aee7e40184156e1151f228f103aa5c6bd0724cc330960f8f" }
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/9b/fb/a70a4214956182e0d7a9099ab17d50bfcba1056188e9b14f35b9e2b62a0d/portalocker-2.10.1-py3-none-any.whl", hash = "sha256:53a5984ebc86a025552264b459b46a2086e269b21823cb572f8f28ee759e45bf" },
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
]
[[package]]
name = "posthog"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "7.7.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
dependencies = [
{ name = "backoff" },
{ name = "distro" },
{ name = "python-dateutil" },
{ name = "requests" },
{ name = "six" },
{ name = "typing-extensions" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/23/dd/ca6d5a79614af27ededc0dca85e77f42f7704e29f8314819d7ce92b9a7f3/posthog-7.7.0.tar.gz", hash = "sha256:b4f2b1a616e099961f6ab61a5a2f88de62082c26801699e556927d21c00737ef" }
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/41/3f/41b426ed9ab161d630edec84bacb6664ae62b6e63af1165919c7e11c17d1/posthog-7.7.0-py3-none-any.whl", hash = "sha256:955f42097bf147459653b9102e5f7f9a22e4b6fc9f15003447bd1137fafbc505" },
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "pot"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "0.9.6.post1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "numpy" },
Bump infinity to v0.6.11. Requires python>=3.11 (#11814) ### What problem does this PR solve? Bump infinity to v0.6.11. Requires python>=3.11 ### Type of change - [x] Refactoring
2025-12-09 16:23:37 +08:00
{ name = "scipy" },
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/42/8b/5f939eaf1fbeb7ff914fe540d659486951a056e5537b8f454362045b6c72/pot-0.9.6.post1.tar.gz", hash = "sha256:9b6cc14a8daecfe1268268168cf46548f9130976b22b24a9e8ec62a734be6c43" }
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/53/17/e4aebb8deef58b0d40ac339d952d12c63559801b50ae43c622d49bebda7e/pot-0.9.6.post1-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:659fff750a162f58b52b33a64c4ac358f4ff44e9dff0841052c088e1b6a54430" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f7/b9/3646c153b13f999ac30112dcf85c5f233af79b0d98c37b52dda9a624c91b/pot-0.9.6.post1-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:4f54830e9f9cb78b1ff7abd5c5bf162625ed6aea903241267c64ea9f0fb73ddb" },
{ url = "https://mirrors.aliyun.com/pypi/packages/53/e9/c7092f7aec8cb32739ad66ba1f1259626546e4893b61b905ce2da3987235/pot-0.9.6.post1-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:e9fd4b1fafacd37debdb984687ddb26f5c43d1429401847d388a6f1bd1f10e98" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0c/a1/f0187ab15aa1538ece07b28f3a7938b8592ef01fbe37b1a8f9c2f8f47f4d/pot-0.9.6.post1-cp313-cp313-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:ec097ec0ef8bb93fee8cdb187b6a0a9653613cba7b06bb603247930e2c629cdc" },
{ url = "https://mirrors.aliyun.com/pypi/packages/29/fa/85af71553b7e990fc37da8d5f2e7294ec66297e62cba419efeec11518e5a/pot-0.9.6.post1-cp313-cp313-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:299f11f172908d799793ef18b2bc82452305350d2528d243e255a17876e98a57" },
{ url = "https://mirrors.aliyun.com/pypi/packages/19/ae/96b2bce173b3d2d3d0faf8b7362fe79e60e1a6a939c9459b2f7b64e625d8/pot-0.9.6.post1-cp313-cp313-win32.whl", hash = "sha256:8a1d95310faae9c75355d9e2fac8dfac41316a2450061eefc982ee498a687a34" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f7/b1/8ca34418e7c4a2ec666e2204539577287223c4e78ab80b1c746cedb559c3/pot-0.9.6.post1-cp313-cp313-win_amd64.whl", hash = "sha256:a43e2b61389bd32f5b488da2488999ed55867e95fedb25dd64f9f390e40b4fab" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
[[package]]
name = "preshed"
version = "3.0.13"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
dependencies = [
{ name = "cymem" },
{ name = "murmurhash" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/43/75/fe6b7bbd0dea530a001b0e24c331b21a0be2786e402abf3c57f5dce43d4b/preshed-3.0.13.tar.gz", hash = "sha256:d75f718bbfd97e992f7827e0fa7faf6a91bdd9c922d5baa4b50d62731396cb89" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/0e/2a/401158195d6dc7f6aef0b354d74d0e95c9da124499448c2b3dbb95b71204/preshed-3.0.13-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:c0d0c14187dc0078d8a63bf190ec045a4d13e7748b6caeb557a7d575e411410b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/88/8f/e20e64573988528785447a6893b2e7ab287ecfd85b3888e978b28812fd20/preshed-3.0.13-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:7770987c2e57497cd26124a9be5f652b5b3ccd0def89859ab0da8bca6144a3de" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b9/72/18168f881359c4482d312f8dc196371bdd61c1583a52b34390da4c88bbea/preshed-3.0.13-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:4a7bc48220de579be6bdb0a8715482cf36e2a625a6fd5ad26c9f43485a4a23b5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fd/3a/3543476091087102775568cea9885dde3453569e9aeee365809108de572f/preshed-3.0.13-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:e5c8462472f790c16708306aef3a102a762bd19dfe3d2f8ee08bd5e12f51b835" },
{ url = "https://mirrors.aliyun.com/pypi/packages/cf/65/b13f01329decc44ef53cfb6b4601ba85382dcb2a4ec78d9250f03a418066/preshed-3.0.13-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:c046736239cc8d72670749b79b526e4111839a2fc461a58545d212797649129c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d1/c7/f1a996c6832234efd4d543041b582418d41ac480ee55c557ec9e65344637/preshed-3.0.13-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:7c333f18e9a81c8a6de0603fd8781e17115324b117c445ca91abdf7bfb1abe49" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e3/b9/96fb71499049885ce19545903fdd38877bbc2be0da47e37c04d01f3e9f66/preshed-3.0.13-cp313-cp313-win_amd64.whl", hash = "sha256:461327f8dd36520dcf1fd55a671e0c3c2c97a2d95e22fc85faa31173f4785dda" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ef/a7/32a4903019d936a2316fdd330bedddac287ac26326107d24fb76a1fbc60a/preshed-3.0.13-cp313-cp313-win_arm64.whl", hash = "sha256:35d6c5acb3ee3b12b87a551913063f0cec784055c2af16e028c19fe875f079d0" },
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "primp"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "0.15.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/56/0b/a87556189da4de1fc6360ca1aa05e8335509633f836cdd06dd17f0743300/primp-0.15.0.tar.gz", hash = "sha256:1af8ea4b15f57571ff7fc5e282a82c5eb69bc695e19b8ddeeda324397965b30a" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/f5/5a/146ac964b99ea7657ad67eb66f770be6577dfe9200cb28f9a95baffd6c3f/primp-0.15.0-cp38-abi3-macosx_10_12_x86_64.whl", hash = "sha256:1b281f4ca41a0c6612d4c6e68b96e28acfe786d226a427cd944baa8d7acd644f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/bc/8a/cc2321e32db3ce64d6e32950d5bcbea01861db97bfb20b5394affc45b387/primp-0.15.0-cp38-abi3-macosx_11_0_arm64.whl", hash = "sha256:489cbab55cd793ceb8f90bb7423c6ea64ebb53208ffcf7a044138e3c66d77299" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c3/7b/cbd5d999a07ff2a21465975d4eb477ae6f69765e8fe8c9087dab250180d8/primp-0.15.0-cp38-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c18b45c23f94016215f62d2334552224236217aaeb716871ce0e4dcfa08eb161" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1b/6e/a6221c612e61303aec2bcac3f0a02e8b67aee8c0db7bdc174aeb8010f975/primp-0.15.0-cp38-abi3-manylinux_2_34_aarch64.whl", hash = "sha256:e985a9cba2e3f96a323722e5440aa9eccaac3178e74b884778e926b5249df080" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3b/54/bfeef5aca613dc660a69d0760a26c6b8747d8fdb5a7f20cb2cee53c9862f/primp-0.15.0-cp38-abi3-manylinux_2_34_armv7l.whl", hash = "sha256:6b84a6ffa083e34668ff0037221d399c24d939b5629cd38223af860de9e17a83" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ac/96/84078e09f16a1dad208f2fe0f8a81be2cf36e024675b0f9eec0c2f6e2182/primp-0.15.0-cp38-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:592f6079646bdf5abbbfc3b0a28dac8de943f8907a250ce09398cda5eaebd260" },
{ url = "https://mirrors.aliyun.com/pypi/packages/6c/80/8a7a9587d3eb85be3d0b64319f2f690c90eb7953e3f73a9ddd9e46c8dc42/primp-0.15.0-cp38-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:5a728e5a05f37db6189eb413d22c78bd143fa59dd6a8a26dacd43332b3971fe8" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0c/dd/f0183ed0145e58cf9d286c1b2c14f63ccee987a4ff79ac85acc31b5d86bd/primp-0.15.0-cp38-abi3-win_amd64.whl", hash = "sha256:aeb6bd20b06dfc92cfe4436939c18de88a58c640752cf7f30d9e4ae893cdec32" },
Feat: Alter flask to Quart for async API serving. (#11275) ### What problem does this PR solve? #11277 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-18 17:05:16 +08:00
]
[[package]]
name = "priority"
version = "2.0.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f5/3c/eb7c35f4dcede96fca1842dac5f4f5d15511aa4b52f3a961219e68ae9204/priority-2.0.0.tar.gz", hash = "sha256:c965d54f1b8d0d0b19479db3924c7c36cf672dbf2aec92d43fbdaf4492ba18c0" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/5e/5f/82c8074f7e84978129347c2c6ec8b6c59f3584ff1a20bc3c940a3e061790/priority-2.0.0-py3-none-any.whl", hash = "sha256:6f8eefce5f3ad59baf2c080a664037bb4725cd0a790d53d59ab4059288faf6aa" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "proces"
version = "0.1.7"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/2c/3d/4159b57736ced0fd22553226df20a985ef7655519c80ffcb8a9fb49ebeee/proces-0.1.7.tar.gz", hash = "sha256:70a05d9e973dd685f7a9092c58be695a8181a411d63796c213232fd3fdc43775" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/6f/88/06cc0c7d890ed8d7e16ef0e56880dea516a21643fb1f3a69a50f4cc6f716/proces-0.1.7-py3-none-any.whl", hash = "sha256:308325bbc96877263f06e57e5e9c760c4b42cc722887ad60be6b18fc37d68762" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "prompt-toolkit"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "3.0.52"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "wcwidth" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a1/96/06e01a7b38dce6fe1db213e061a4602dd6032a8a97ef6c1a862537732421/prompt_toolkit-3.0.52.tar.gz", hash = "sha256:28cde192929c8e7321de85de1ddbe736f1375148b02f2e17edd840042b1be855" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/84/03/0d3ce49e2505ae70cf43bc5bb3033955d2fc9f932163e84dc0779cc47f48/prompt_toolkit-3.0.52-py3-none-any.whl", hash = "sha256:9aac639a3bbd33284347de5ad8d68ecc044b91a762dc39b7c21095fcd6a19955" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "propcache"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "0.4.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/9e/da/e9fc233cf63743258bff22b3dfa7ea5baef7b5bc324af47a0ad89b8ffc6f/propcache-0.4.1.tar.gz", hash = "sha256:f48107a8c637e80362555f37ecf49abe20370e557cc4ab374f04ec4423c97c3d" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/bf/df/6d9c1b6ac12b003837dde8a10231a7344512186e87b36e855bef32241942/propcache-0.4.1-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:43eedf29202c08550aac1d14e0ee619b0430aaef78f85864c1a892294fbc28cf" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8b/e8/677a0025e8a2acf07d3418a2e7ba529c9c33caf09d3c1f25513023c1db56/propcache-0.4.1-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:d62cdfcfd89ccb8de04e0eda998535c406bf5e060ffd56be6c586cbcc05b3311" },
{ url = "https://mirrors.aliyun.com/pypi/packages/89/a4/92380f7ca60f99ebae761936bc48a72a639e8a47b29050615eef757cb2a7/propcache-0.4.1-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:cae65ad55793da34db5f54e4029b89d3b9b9490d8abe1b4c7ab5d4b8ec7ebf74" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2d/48/c5ac64dee5262044348d1d78a5f85dd1a57464a60d30daee946699963eb3/propcache-0.4.1-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:333ddb9031d2704a301ee3e506dc46b1fe5f294ec198ed6435ad5b6a085facfe" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c6/0c/cd762dd011a9287389a6a3eb43aa30207bde253610cca06824aeabfe9653/propcache-0.4.1-cp313-cp313-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:fd0858c20f078a32cf55f7e81473d96dcf3b93fd2ccdb3d40fdf54b8573df3af" },
{ url = "https://mirrors.aliyun.com/pypi/packages/30/3e/49861e90233ba36890ae0ca4c660e95df565b2cd15d4a68556ab5865974e/propcache-0.4.1-cp313-cp313-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:678ae89ebc632c5c204c794f8dab2837c5f159aeb59e6ed0539500400577298c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f1/8b/544bc867e24e1bd48f3118cecd3b05c694e160a168478fa28770f22fd094/propcache-0.4.1-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:d472aeb4fbf9865e0c6d622d7f4d54a4e101a89715d8904282bb5f9a2f476c3f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/50/a6/4282772fd016a76d3e5c0df58380a5ea64900afd836cec2c2f662d1b9bb3/propcache-0.4.1-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:4d3df5fa7e36b3225954fba85589da77a0fe6a53e3976de39caf04a0db4c36f1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3e/ec/d8a7cd406ee1ddb705db2139f8a10a8a427100347bd698e7014351c7af09/propcache-0.4.1-cp313-cp313-musllinux_1_2_armv7l.whl", hash = "sha256:ee17f18d2498f2673e432faaa71698032b0127ebf23ae5974eeaf806c279df24" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f6/6c/f38ab64af3764f431e359f8baf9e0a21013e24329e8b85d2da32e8ed07ca/propcache-0.4.1-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:580e97762b950f993ae618e167e7be9256b8353c2dcd8b99ec100eb50f5286aa" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d6/e3/fa846bd70f6534d647886621388f0a265254d30e3ce47e5c8e6e27dbf153/propcache-0.4.1-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:501d20b891688eb8e7aa903021f0b72d5a55db40ffaab27edefd1027caaafa61" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e2/39/8163fc6f3133fea7b5f2827e8eba2029a0277ab2c5beee6c1db7b10fc23d/propcache-0.4.1-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:9a0bd56e5b100aef69bd8562b74b46254e7c8812918d3baa700c8a8009b0af66" },
{ url = "https://mirrors.aliyun.com/pypi/packages/93/89/caa9089970ca49c7c01662bd0eeedfe85494e863e8043565aeb6472ce8fe/propcache-0.4.1-cp313-cp313-win32.whl", hash = "sha256:bcc9aaa5d80322bc2fb24bb7accb4a30f81e90ab8d6ba187aec0744bc302ad81" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f5/ab/f76ec3c3627c883215b5c8080debb4394ef5a7a29be811f786415fc1e6fd/propcache-0.4.1-cp313-cp313-win_amd64.whl", hash = "sha256:381914df18634f5494334d201e98245c0596067504b9372d8cf93f4bb23e025e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/59/1b/e71ae98235f8e2ba5004d8cb19765a74877abf189bc53fc0c80d799e56c3/propcache-0.4.1-cp313-cp313-win_arm64.whl", hash = "sha256:8873eb4460fd55333ea49b7d189749ecf6e55bf85080f11b1c4530ed3034cba1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/83/ce/a31bbdfc24ee0dcbba458c8175ed26089cf109a55bbe7b7640ed2470cfe9/propcache-0.4.1-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:92d1935ee1f8d7442da9c0c4fa7ac20d07e94064184811b685f5c4fada64553b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/25/9c/442a45a470a68456e710d96cacd3573ef26a1d0a60067e6a7d5e655621ed/propcache-0.4.1-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:473c61b39e1460d386479b9b2f337da492042447c9b685f28be4f74d3529e566" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f4/bf/b1d5e21dbc3b2e889ea4327044fb16312a736d97640fb8b6aa3f9c7b3b65/propcache-0.4.1-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:c0ef0aaafc66fbd87842a3fe3902fd889825646bc21149eafe47be6072725835" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f4/04/5b4c54a103d480e978d3c8a76073502b18db0c4bc17ab91b3cb5092ad949/propcache-0.4.1-cp313-cp313t-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:f95393b4d66bfae908c3ca8d169d5f79cd65636ae15b5e7a4f6e67af675adb0e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b4/c1/86f846827fb969c4b78b0af79bba1d1ea2156492e1b83dea8b8a6ae27395/propcache-0.4.1-cp313-cp313t-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:c07fda85708bc48578467e85099645167a955ba093be0a2dcba962195676e859" },
{ url = "https://mirrors.aliyun.com/pypi/packages/36/1d/fc272a63c8d3bbad6878c336c7a7dea15e8f2d23a544bda43205dfa83ada/propcache-0.4.1-cp313-cp313t-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:af223b406d6d000830c6f65f1e6431783fc3f713ba3e6cc8c024d5ee96170a4b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/07/0c/01f2219d39f7e53d52e5173bcb09c976609ba30209912a0680adfb8c593a/propcache-0.4.1-cp313-cp313t-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:a78372c932c90ee474559c5ddfffd718238e8673c340dc21fe45c5b8b54559a0" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2d/18/cd28081658ce597898f0c4d174d4d0f3c5b6d4dc27ffafeef835c95eb359/propcache-0.4.1-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:564d9f0d4d9509e1a870c920a89b2fec951b44bf5ba7d537a9e7c1ccec2c18af" },
{ url = "https://mirrors.aliyun.com/pypi/packages/7a/71/1f9e22eb8b8316701c2a19fa1f388c8a3185082607da8e406a803c9b954e/propcache-0.4.1-cp313-cp313t-musllinux_1_2_armv7l.whl", hash = "sha256:17612831fda0138059cc5546f4d12a2aacfb9e47068c06af35c400ba58ba7393" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4a/65/3d4b61f36af2b4eddba9def857959f1016a51066b4f1ce348e0cf7881f58/propcache-0.4.1-cp313-cp313t-musllinux_1_2_ppc64le.whl", hash = "sha256:41a89040cb10bd345b3c1a873b2bf36413d48da1def52f268a055f7398514874" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2a/42/26746ab087faa77c1c68079b228810436ccd9a5ce9ac85e2b7307195fd06/propcache-0.4.1-cp313-cp313t-musllinux_1_2_s390x.whl", hash = "sha256:e35b88984e7fa64aacecea39236cee32dd9bd8c55f57ba8a75cf2399553f9bd7" },
{ url = "https://mirrors.aliyun.com/pypi/packages/94/13/630690fe201f5502d2403dd3cfd451ed8858fe3c738ee88d095ad2ff407b/propcache-0.4.1-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:6f8b465489f927b0df505cbe26ffbeed4d6d8a2bbc61ce90eb074ff129ef0ab1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/92/f7/1d4ec5841505f423469efbfc381d64b7b467438cd5a4bbcbb063f3b73d27/propcache-0.4.1-cp313-cp313t-win32.whl", hash = "sha256:2ad890caa1d928c7c2965b48f3a3815c853180831d0e5503d35cf00c472f4717" },
{ url = "https://mirrors.aliyun.com/pypi/packages/48/f0/615c30622316496d2cbbc29f5985f7777d3ada70f23370608c1d3e081c1f/propcache-0.4.1-cp313-cp313t-win_amd64.whl", hash = "sha256:f7ee0e597f495cf415bcbd3da3caa3bd7e816b74d0d52b8145954c5e6fd3ff37" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fd/ca/6002e46eccbe0e33dcd4069ef32f7f1c9e243736e07adca37ae8c4830ec3/propcache-0.4.1-cp313-cp313t-win_arm64.whl", hash = "sha256:929d7cbe1f01bb7baffb33dc14eb5691c95831450a26354cd210a8155170c93a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/5b/5a/bc7b4a4ef808fa59a816c17b20c4bef6884daebbdf627ff2a161da67da19/propcache-0.4.1-py3-none-any.whl", hash = "sha256:af2a6052aeb6cf17d3e46ee169099044fd8224cbaf75c76a2ef596e8163e2237" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "proto-plus"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.27.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "protobuf" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/01/89/9cbe2f4bba860e149108b683bc2efec21f14d5f7ed6e25562ad86acbc373/proto_plus-1.27.0.tar.gz", hash = "sha256:873af56dd0d7e91836aee871e5799e1c6f1bda86ac9a983e0bb9f0c266a568c4" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/cd/24/3b7a0818484df9c28172857af32c2397b6d8fcd99d9468bd4684f98ebf0a/proto_plus-1.27.0-py3-none-any.whl", hash = "sha256:1baa7f81cf0f8acb8bc1f6d085008ba4171eaf669629d1b6d1673b21ed1c0a82" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "protobuf"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "5.29.5"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/43/29/d09e70352e4e88c9c7a198d5645d7277811448d76c23b00345670f7c8a38/protobuf-5.29.5.tar.gz", hash = "sha256:bc1463bafd4b0929216c35f437a8e28731a2b7fe3d98bb77a600efced5a15c84" }
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/5f/11/6e40e9fc5bba02988a214c07cf324595789ca7820160bfd1f8be96e48539/protobuf-5.29.5-cp310-abi3-win32.whl", hash = "sha256:3f1c6468a2cfd102ff4703976138844f78ebd1fb45f49011afc5139e9e283079" },
{ url = "https://mirrors.aliyun.com/pypi/packages/81/7f/73cefb093e1a2a7c3ffd839e6f9fcafb7a427d300c7f8aef9c64405d8ac6/protobuf-5.29.5-cp310-abi3-win_amd64.whl", hash = "sha256:3f76e3a3675b4a4d867b52e4a5f5b78a2ef9565549d4037e06cf7b0942b1d3fc" },
{ url = "https://mirrors.aliyun.com/pypi/packages/dd/73/10e1661c21f139f2c6ad9b23040ff36fee624310dc28fba20d33fdae124c/protobuf-5.29.5-cp38-abi3-macosx_10_9_universal2.whl", hash = "sha256:e38c5add5a311f2a6eb0340716ef9b039c1dfa428b28f25a7838ac329204a671" },
{ url = "https://mirrors.aliyun.com/pypi/packages/6c/04/98f6f8cf5b07ab1294c13f34b4e69b3722bb609c5b701d6c169828f9f8aa/protobuf-5.29.5-cp38-abi3-manylinux2014_aarch64.whl", hash = "sha256:fa18533a299d7ab6c55a238bf8629311439995f2e7eca5caaff08663606e9015" },
{ url = "https://mirrors.aliyun.com/pypi/packages/85/e4/07c80521879c2d15f321465ac24c70efe2381378c00bf5e56a0f4fbac8cd/protobuf-5.29.5-cp38-abi3-manylinux2014_x86_64.whl", hash = "sha256:63848923da3325e1bf7e9003d680ce6e14b07e55d0473253a690c3a8b8fd6e61" },
{ url = "https://mirrors.aliyun.com/pypi/packages/7e/cc/7e77861000a0691aeea8f4566e5d3aa716f2b1dece4a24439437e41d3d25/protobuf-5.29.5-py3-none-any.whl", hash = "sha256:6cf42630262c59b2d8de33954443d94b746c952b01434fc58a417fdbd2e84bd5" },
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
]
[[package]]
name = "psutil"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "7.2.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/73/cb/09e5184fb5fc0358d110fc3ca7f6b1d033800734d34cac10f4136cfac10e/psutil-7.2.1.tar.gz", hash = "sha256:f7583aec590485b43ca601dd9cea0dcd65bd7bb21d30ef4ddbf4ea6b5ed1bdd3" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/77/8e/f0c242053a368c2aa89584ecd1b054a18683f13d6e5a318fc9ec36582c94/psutil-7.2.1-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:ba9f33bb525b14c3ea563b2fd521a84d2fa214ec59e3e6a2858f78d0844dd60d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/26/97/a58a4968f8990617decee234258a2b4fc7cd9e35668387646c1963e69f26/psutil-7.2.1-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:81442dac7abfc2f4f4385ea9e12ddf5a796721c0f6133260687fec5c3780fa49" },
{ url = "https://mirrors.aliyun.com/pypi/packages/db/6d/ed44901e830739af5f72a85fa7ec5ff1edea7f81bfbf4875e409007149bd/psutil-7.2.1-cp313-cp313t-manylinux2010_x86_64.manylinux_2_12_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:ea46c0d060491051d39f0d2cff4f98d5c72b288289f57a21556cc7d504db37fc" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c7/65/b628f8459bca4efbfae50d4bf3feaab803de9a160b9d5f3bd9295a33f0c2/psutil-7.2.1-cp313-cp313t-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:35630d5af80d5d0d49cfc4d64c1c13838baf6717a13effb35869a5919b854cdf" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fb/23/851cadc9764edcc18f0effe7d0bf69f727d4cf2442deb4a9f78d4e4f30f2/psutil-7.2.1-cp313-cp313t-win_amd64.whl", hash = "sha256:923f8653416604e356073e6e0bccbe7c09990acef442def2f5640dd0faa9689f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/59/82/d63e8494ec5758029f31c6cb06d7d161175d8281e91d011a4a441c8a43b5/psutil-7.2.1-cp313-cp313t-win_arm64.whl", hash = "sha256:cfbe6b40ca48019a51827f20d830887b3107a74a79b01ceb8cc8de4ccb17b672" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c5/cf/5180eb8c8bdf6a503c6919f1da28328bd1e6b3b1b5b9d5b01ae64f019616/psutil-7.2.1-cp36-abi3-macosx_10_9_x86_64.whl", hash = "sha256:b2e953fcfaedcfbc952b44744f22d16575d3aa78eb4f51ae74165b4e96e55f42" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c5/2c/78e4a789306a92ade5000da4f5de3255202c534acdadc3aac7b5458fadef/psutil-7.2.1-cp36-abi3-macosx_11_0_arm64.whl", hash = "sha256:05cc68dbb8c174828624062e73078e7e35406f4ca2d0866c272c2410d8ef06d1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/29/f8/40e01c350ad9a2b3cb4e6adbcc8a83b17ee50dd5792102b6142385937db5/psutil-7.2.1-cp36-abi3-manylinux2010_x86_64.manylinux_2_12_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:5e38404ca2bb30ed7267a46c02f06ff842e92da3bb8c5bfdadbd35a5722314d8" },
{ url = "https://mirrors.aliyun.com/pypi/packages/06/e4/b751cdf839c011a9714a783f120e6a86b7494eb70044d7d81a25a5cd295f/psutil-7.2.1-cp36-abi3-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:ab2b98c9fc19f13f59628d94df5cc4cc4844bc572467d113a8b517d634e362c6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/44/ad/bbf6595a8134ee1e94a4487af3f132cef7fce43aef4a93b49912a48c3af7/psutil-7.2.1-cp36-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:f78baafb38436d5a128f837fab2d92c276dfb48af01a240b861ae02b2413ada8" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1c/15/dd6fd869753ce82ff64dcbc18356093471a5a5adf4f77ed1f805d473d859/psutil-7.2.1-cp36-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:99a4cd17a5fdd1f3d014396502daa70b5ec21bf4ffe38393e152f8e449757d67" },
{ url = "https://mirrors.aliyun.com/pypi/packages/34/68/d9317542e3f2b180c4306e3f45d3c922d7e86d8ce39f941bb9e2e9d8599e/psutil-7.2.1-cp37-abi3-win_amd64.whl", hash = "sha256:b1b0671619343aa71c20ff9767eced0483e4fc9e1f489d50923738caf6a03c17" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3e/73/2ce007f4198c80fcf2cb24c169884f833fe93fbc03d55d302627b094ee91/psutil-7.2.1-cp37-abi3-win_arm64.whl", hash = "sha256:0d67c1822c355aa6f7314d92018fb4268a76668a536f133599b91edd48759442" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "psycopg2-binary"
Bump python to >=3.12 (#11846) ### What problem does this PR solve? Bump python to >=3.12 ### Type of change - [x] Refactoring
2025-12-09 19:55:25 +08:00
version = "2.9.11"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ac/6c/8767aaa597ba424643dc87348c6f1754dd9f48e80fdc1b9f7ca5c3a7c213/psycopg2-binary-2.9.11.tar.gz", hash = "sha256:b6aed9e096bf63f9e75edf2581aa9a7e7186d97ab5c177aa6c87797cd591236c" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ff/a8/a2709681b3ac11b0b1786def10006b8995125ba268c9a54bea6f5ae8bd3e/psycopg2_binary-2.9.11-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:b8fb3db325435d34235b044b199e56cdf9ff41223a4b9752e8576465170bb38c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/62/e1/c2b38d256d0dafd32713e9f31982a5b028f4a3651f446be70785f484f472/psycopg2_binary-2.9.11-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:366df99e710a2acd90efed3764bb1e28df6c675d33a7fb40df9b7281694432ee" },
{ url = "https://mirrors.aliyun.com/pypi/packages/11/32/b2ffe8f3853c181e88f0a157c5fb4e383102238d73c52ac6d93a5c8bffe6/psycopg2_binary-2.9.11-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:8c55b385daa2f92cb64b12ec4536c66954ac53654c7f15a203578da4e78105c0" },
{ url = "https://mirrors.aliyun.com/pypi/packages/10/04/6ca7477e6160ae258dc96f67c371157776564679aefd247b66f4661501a2/psycopg2_binary-2.9.11-cp313-cp313-manylinux2014_ppc64le.manylinux_2_17_ppc64le.whl", hash = "sha256:c0377174bf1dd416993d16edc15357f6eb17ac998244cca19bc67cdc0e2e5766" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3c/7e/6a1a38f86412df101435809f225d57c1a021307dd0689f7a5e7fe83588b1/psycopg2_binary-2.9.11-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:5c6ff3335ce08c75afaed19e08699e8aacf95d4a260b495a4a8545244fe2ceb3" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f2/7d/c07374c501b45f3579a9eb761cbf2604ddef3d96ad48679112c2c5aa9c25/psycopg2_binary-2.9.11-cp313-cp313-manylinux_2_38_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:84011ba3109e06ac412f95399b704d3d6950e386b7994475b231cf61eec2fc1f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/82/56/993b7104cb8345ad7d4516538ccf8f0d0ac640b1ebd8c754a7b024e76878/psycopg2_binary-2.9.11-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:ba34475ceb08cccbdd98f6b46916917ae6eeb92b5ae111df10b544c3a4621dc4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2d/ac/eaeb6029362fd8d454a27374d84c6866c82c33bfc24587b4face5a8e43ef/psycopg2_binary-2.9.11-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:b31e90fdd0f968c2de3b26ab014314fe814225b6c324f770952f7d38abf17e3c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2b/39/50c3facc66bded9ada5cbc0de867499a703dc6bca6be03070b4e3b65da6c/psycopg2_binary-2.9.11-cp313-cp313-musllinux_1_2_riscv64.whl", hash = "sha256:d526864e0f67f74937a8fce859bd56c979f5e2ec57ca7c627f5f1071ef7fee60" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9c/8e/b7de019a1f562f72ada81081a12823d3c1590bedc48d7d2559410a2763fe/psycopg2_binary-2.9.11-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:04195548662fa544626c8ea0f06561eb6203f1984ba5b4562764fbeb4c3d14b1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/80/2d/1bb683f64737bbb1f86c82b7359db1eb2be4e2c0c13b947f80efefa7d3e5/psycopg2_binary-2.9.11-cp313-cp313-win_amd64.whl", hash = "sha256:efff12b432179443f54e230fdf60de1f6cc726b6c832db8701227d089310e8aa" },
Bump python to >=3.12 (#11846) ### What problem does this PR solve? Bump python to >=3.12 ### Type of change - [x] Refactoring
2025-12-09 19:55:25 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "py"
version = "1.11.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/98/ff/fec109ceb715d2a6b4c4a85a61af3b40c723a961e8828319fbcb15b868dc/py-1.11.0.tar.gz", hash = "sha256:51c75c4126074b472f746a24399ad32f6053d1b34b68d2fa41e558e6f4a98719" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/f6/f0/10642828a8dfb741e5f3fbaac830550a518a775c7fff6f04a007259b0548/py-1.11.0-py2.py3-none-any.whl", hash = "sha256:607c53218732647dff4acdfcd50cb62615cedf612e72d1724fb1a0cc6405b378" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "py-mini-racer"
version = "0.6.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/50/97/a578b918b2e5923dd754cb60bb8b8aeffc85255ffb92566e3c65b148ff72/py_mini_racer-0.6.0.tar.gz", hash = "sha256:f71e36b643d947ba698c57cd9bd2232c83ca997b0802fc2f7f79582377040c11" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/29/a9/8ce0ca222ef04d602924a1e099be93f5435ca6f3294182a30574d4159ca2/py_mini_racer-0.6.0-py2.py3-none-manylinux1_x86_64.whl", hash = "sha256:42896c24968481dd953eeeb11de331f6870917811961c9b26ba09071e07180e2" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: add Airtable connector and integration for data synchronization (#12211) ### What problem does this PR solve? change: add Airtable connector and integration for data synchronization ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-25 17:50:41 +08:00
[[package]]
name = "pyairtable"
version = "3.3.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: add Airtable connector and integration for data synchronization (#12211) ### What problem does this PR solve? change: add Airtable connector and integration for data synchronization ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-25 17:50:41 +08:00
dependencies = [
{ name = "inflection" },
{ name = "pydantic" },
{ name = "requests" },
{ name = "typing-extensions" },
{ name = "urllib3" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/2c/1d/8a572580e02297cef7ae01053a8b550b7759ea80326cd3231df87b00555b/pyairtable-3.3.0.tar.gz", hash = "sha256:d6d3b77f6feb7a02a84779c2235d37a46605f36030cf20ed99b08bab73108a8c" }
Feat: add Airtable connector and integration for data synchronization (#12211) ### What problem does this PR solve? change: add Airtable connector and integration for data synchronization ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-25 17:50:41 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/13/7b/bebb0ebb86353b63740869ed10ac1fef1636ccc6042beb1d8d3956cad02d/pyairtable-3.3.0-py2.py3-none-any.whl", hash = "sha256:38af09c18659918b96539ac4d9730c9656f6ce2088cdff692dd311fa16802acf" },
Feat: add Airtable connector and integration for data synchronization (#12211) ### What problem does this PR solve? change: add Airtable connector and integration for data synchronization ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-25 17:50:41 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "pyarrow"
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
version = "22.0.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/30/53/04a7fdc63e6056116c9ddc8b43bc28c12cdd181b85cbeadb79278475f3ae/pyarrow-22.0.0.tar.gz", hash = "sha256:3d600dc583260d845c7d8a6db540339dd883081925da2bd1c5cb808f720b3cd9" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/a6/d6/d0fac16a2963002fc22c8fa75180a838737203d558f0ed3b564c4a54eef5/pyarrow-22.0.0-cp313-cp313-macosx_12_0_arm64.whl", hash = "sha256:e6e95176209257803a8b3d0394f21604e796dadb643d2f7ca21b66c9c0b30c9a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c6/9c/1d6357347fbae062ad3f17082f9ebc29cc733321e892c0d2085f42a2212b/pyarrow-22.0.0-cp313-cp313-macosx_12_0_x86_64.whl", hash = "sha256:001ea83a58024818826a9e3f89bf9310a114f7e26dfe404a4c32686f97bd7901" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ff/c0/782344c2ce58afbea010150df07e3a2f5fdad299cd631697ae7bd3bac6e3/pyarrow-22.0.0-cp313-cp313-manylinux_2_28_aarch64.whl", hash = "sha256:ce20fe000754f477c8a9125543f1936ea5b8867c5406757c224d745ed033e691" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1b/8b/5362443737a5307a7b67c1017c42cd104213189b4970bf607e05faf9c525/pyarrow-22.0.0-cp313-cp313-manylinux_2_28_x86_64.whl", hash = "sha256:e0a15757fccb38c410947df156f9749ae4a3c89b2393741a50521f39a8cf202a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/69/4d/76e567a4fc2e190ee6072967cb4672b7d9249ac59ae65af2d7e3047afa3b/pyarrow-22.0.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:cedb9dd9358e4ea1d9bce3665ce0797f6adf97ff142c8e25b46ba9cdd508e9b6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/01/5e/5653f0535d2a1aef8223cee9d92944cb6bccfee5cf1cd3f462d7cb022790/pyarrow-22.0.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:252be4a05f9d9185bb8c18e83764ebcfea7185076c07a7a662253af3a8c07941" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2d/f8/1d0bd75bf9328a3b826e24a16e5517cd7f9fbf8d34a3184a4566ef5a7f29/pyarrow-22.0.0-cp313-cp313-win_amd64.whl", hash = "sha256:a4893d31e5ef780b6edcaf63122df0f8d321088bb0dee4c8c06eccb1ca28d145" },
{ url = "https://mirrors.aliyun.com/pypi/packages/90/81/db56870c997805bf2b0f6eeeb2d68458bf4654652dccdcf1bf7a42d80903/pyarrow-22.0.0-cp313-cp313t-macosx_12_0_arm64.whl", hash = "sha256:f7fe3dbe871294ba70d789be16b6e7e52b418311e166e0e3cba9522f0f437fb1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1c/98/0727947f199aba8a120f47dfc229eeb05df15bcd7a6f1b669e9f882afc58/pyarrow-22.0.0-cp313-cp313t-macosx_12_0_x86_64.whl", hash = "sha256:ba95112d15fd4f1105fb2402c4eab9068f0554435e9b7085924bcfaac2cc306f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/96/b4/9babdef9c01720a0785945c7cf550e4acd0ebcd7bdd2e6f0aa7981fa85e2/pyarrow-22.0.0-cp313-cp313t-manylinux_2_28_aarch64.whl", hash = "sha256:c064e28361c05d72eed8e744c9605cbd6d2bb7481a511c74071fd9b24bc65d7d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f8/ca/2f8804edd6279f78a37062d813de3f16f29183874447ef6d1aadbb4efa0f/pyarrow-22.0.0-cp313-cp313t-manylinux_2_28_x86_64.whl", hash = "sha256:6f9762274496c244d951c819348afbcf212714902742225f649cf02823a6a10f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b9/f0/77aa5198fd3943682b2e4faaf179a674f0edea0d55d326d83cb2277d9363/pyarrow-22.0.0-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:a9d9ffdc2ab696f6b15b4d1f7cec6658e1d788124418cb30030afbae31c64746" },
{ url = "https://mirrors.aliyun.com/pypi/packages/79/87/a1937b6e78b2aff18b706d738c9e46ade5bfcf11b294e39c87706a0089ac/pyarrow-22.0.0-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:ec1a15968a9d80da01e1d30349b2b0d7cc91e96588ee324ce1b5228175043e95" },
{ url = "https://mirrors.aliyun.com/pypi/packages/60/ae/b5a5811e11f25788ccfdaa8f26b6791c9807119dffcf80514505527c384c/pyarrow-22.0.0-cp313-cp313t-win_amd64.whl", hash = "sha256:bba208d9c7decf9961998edf5c65e3ea4355d5818dd6cd0f6809bec1afb951cc" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "pyasn1"
Security: do not use litellm 1.82.7 and 1.82.8 (#13768) ### What problem does this PR solve? See [issue](https://github.com/BerriAI/litellm/issues/24518) from Litellm. Upgraded from `1.81.15` to `1.82.6`, so RAGFlow is safe as always. ### Type of change - [x] Security Co-authored-by: Jin Hai <haijin.chn@gmail.com> Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
2026-03-25 22:39:33 +08:00
version = "0.6.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/5c/5f/6583902b6f79b399c9c40674ac384fd9cd77805f9e6205075f828ef11fb2/pyasn1-0.6.3.tar.gz", hash = "sha256:697a8ecd6d98891189184ca1fa05d1bb00e2f84b5977c481452050549c8a72cf" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/5d/a0/7d793dce3fa811fe047d6ae2431c672364b462850c6235ae306c0efd025f/pyasn1-0.6.3-py3-none-any.whl", hash = "sha256:a80184d120f0864a52a073acc6fc642847d0be408e7c7252f31390c0f4eadcde" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "pyasn1-modules"
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
version = "0.4.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "pyasn1" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/e9/e6/78ebbb10a8c8e4b61a59249394a4a594c1a7af95593dc933a349c8d00964/pyasn1_modules-0.4.2.tar.gz", hash = "sha256:677091de870a80aae844b1ca6134f54652fa2c8c5a52aa396440ac3106e941e6" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/47/8d/d529b5d697919ba8c11ad626e835d4039be708a35b0d22de83a269a6682c/pyasn1_modules-0.4.2-py3-none-any.whl", hash = "sha256:29253a9207ce32b64c3ac6600edc75368f98473906e8fd1043bd6b5b1de2c14a" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "pyclipper"
Bump python to >=3.12 (#11846) ### What problem does this PR solve? Bump python to >=3.12 ### Type of change - [x] Refactoring
2025-12-09 19:55:25 +08:00
version = "1.4.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f6/21/3c06205bb407e1f79b73b7b4dfb3950bd9537c4f625a68ab5cc41177f5bc/pyclipper-1.4.0.tar.gz", hash = "sha256:9882bd889f27da78add4dd6f881d25697efc740bf840274e749988d25496c8e1" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/67/d0/cbce7d47de1e6458f66a4d999b091640134deb8f2c7351eab993b70d2e10/pyclipper-1.4.0-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:d49df13cbb2627ccb13a1046f3ea6ebf7177b5504ec61bdef87d6a704046fd6e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ce/cc/742b9d69d96c58ac156947e1b56d0f81cbacbccf869e2ac7229f2f86dc4e/pyclipper-1.4.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:37bfec361e174110cdddffd5ecd070a8064015c99383d95eb692c253951eee8a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/db/48/dd301d62c1529efdd721b47b9e5fb52120fcdac5f4d3405cfc0d2f391414/pyclipper-1.4.0-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:14c8bdb5a72004b721c4e6f448d2c2262d74a7f0c9e3076aeff41e564a92389f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/07/bf/d493fd1b33bb090fa64e28c1009374d5d72fa705f9331cd56517c35e381e/pyclipper-1.4.0-cp313-cp313-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:f2a50c22c3a78cb4e48347ecf06930f61ce98cf9252f2e292aa025471e9d75b1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/cf/88/b95ea8ea21ddca34aa14b123226a81526dd2faaa993f9aabd3ed21231604/pyclipper-1.4.0-cp313-cp313-win32.whl", hash = "sha256:c9a3faa416ff536cee93417a72bfb690d9dea136dc39a39dbbe1e5dadf108c9c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ba/42/0a1920d276a0e1ca21dc0d13ee9e3ba10a9a8aa3abac76cd5e5a9f503306/pyclipper-1.4.0-cp313-cp313-win_amd64.whl", hash = "sha256:d4b2d7c41086f1927d14947c563dfc7beed2f6c0d9af13c42fe3dcdc20d35832" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "pycparser"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "2.23"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/fe/cf/d2d3b9f5699fb1e4615c8e32ff220203e43b248e1dfcc6736ad9057731ca/pycparser-2.23.tar.gz", hash = "sha256:78816d4f24add8f10a06d6f05b4d424ad9e96cfebf68a4ddc99c65c0720d00c2" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/a0/e3/59cd50310fc9b59512193629e1984c1f95e5c8ae6e5d8c69532ccc65a7fe/pycparser-2.23-py3-none-any.whl", hash = "sha256:e5c6e8d3fbad53479cab09ac03729e0a9faf2bee3db8208a550daf5af81a5934" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "pycryptodome"
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
version = "3.23.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/8e/a6/8452177684d5e906854776276ddd34eca30d1b1e15aa1ee9cefc289a33f5/pycryptodome-3.23.0.tar.gz", hash = "sha256:447700a657182d60338bab09fdb27518f8856aecd80ae4c6bdddb67ff5da44ef" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/04/5d/bdb09489b63cd34a976cc9e2a8d938114f7a53a74d3dd4f125ffa49dce82/pycryptodome-3.23.0-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:0011f7f00cdb74879142011f95133274741778abba114ceca229adbf8e62c3e4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a7/ce/7840250ed4cc0039c433cd41715536f926d6e86ce84e904068eb3244b6a6/pycryptodome-3.23.0-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:90460fc9e088ce095f9ee8356722d4f10f86e5be06e2354230a9880b9c549aae" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ee/f0/991da24c55c1f688d6a3b5a11940567353f74590734ee4a64294834ae472/pycryptodome-3.23.0-cp313-cp313t-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4764e64b269fc83b00f682c47443c2e6e85b18273712b98aa43bcb77f8570477" },
{ url = "https://mirrors.aliyun.com/pypi/packages/54/16/0e11882deddf00f68b68dd4e8e442ddc30641f31afeb2bc25588124ac8de/pycryptodome-3.23.0-cp313-cp313t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:eb8f24adb74984aa0e5d07a2368ad95276cf38051fe2dc6605cbcf482e04f2a7" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d5/fc/4347fea23a3f95ffb931f383ff28b3f7b1fe868739182cb76718c0da86a1/pycryptodome-3.23.0-cp313-cp313t-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:d97618c9c6684a97ef7637ba43bdf6663a2e2e77efe0f863cce97a76af396446" },
{ url = "https://mirrors.aliyun.com/pypi/packages/6e/d9/c5261780b69ce66d8cfab25d2797bd6e82ba0241804694cd48be41add5eb/pycryptodome-3.23.0-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:9a53a4fe5cb075075d515797d6ce2f56772ea7e6a1e5e4b96cf78a14bac3d265" },
{ url = "https://mirrors.aliyun.com/pypi/packages/5a/6f/3af2ffedd5cfa08c631f89452c6648c4d779e7772dfc388c77c920ca6bbf/pycryptodome-3.23.0-cp313-cp313t-musllinux_1_2_i686.whl", hash = "sha256:763d1d74f56f031788e5d307029caef067febf890cd1f8bf61183ae142f1a77b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9a/dc/9060d807039ee5de6e2f260f72f3d70ac213993a804f5e67e0a73a56dd2f/pycryptodome-3.23.0-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:954af0e2bd7cea83ce72243b14e4fb518b18f0c1649b576d114973e2073b273d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f9/34/e6c8ca177cb29dcc4967fef73f5de445912f93bd0343c9c33c8e5bf8cde8/pycryptodome-3.23.0-cp313-cp313t-win32.whl", hash = "sha256:257bb3572c63ad8ba40b89f6fc9d63a2a628e9f9708d31ee26560925ebe0210a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e4/1d/89756b8d7ff623ad0160f4539da571d1f594d21ee6d68be130a6eccb39a4/pycryptodome-3.23.0-cp313-cp313t-win_amd64.whl", hash = "sha256:6501790c5b62a29fcb227bd6b62012181d886a767ce9ed03b303d1f22eb5c625" },
{ url = "https://mirrors.aliyun.com/pypi/packages/5d/61/35a64f0feaea9fd07f0d91209e7be91726eb48c0f1bfc6720647194071e4/pycryptodome-3.23.0-cp313-cp313t-win_arm64.whl", hash = "sha256:9a77627a330ab23ca43b48b130e202582e91cc69619947840ea4d2d1be21eb39" },
{ url = "https://mirrors.aliyun.com/pypi/packages/db/6c/a1f71542c969912bb0e106f64f60a56cc1f0fabecf9396f45accbe63fa68/pycryptodome-3.23.0-cp37-abi3-macosx_10_9_universal2.whl", hash = "sha256:187058ab80b3281b1de11c2e6842a357a1f71b42cb1e15bce373f3d238135c27" },
{ url = "https://mirrors.aliyun.com/pypi/packages/6e/4e/a066527e079fc5002390c8acdd3aca431e6ea0a50ffd7201551175b47323/pycryptodome-3.23.0-cp37-abi3-macosx_10_9_x86_64.whl", hash = "sha256:cfb5cd445280c5b0a4e6187a7ce8de5a07b5f3f897f235caa11f1f435f182843" },
{ url = "https://mirrors.aliyun.com/pypi/packages/50/52/adaf4c8c100a8c49d2bd058e5b551f73dfd8cb89eb4911e25a0c469b6b4e/pycryptodome-3.23.0-cp37-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:67bd81fcbe34f43ad9422ee8fd4843c8e7198dd88dd3d40e6de42ee65fbe1490" },
{ url = "https://mirrors.aliyun.com/pypi/packages/5f/e9/a09476d436d0ff1402ac3867d933c61805ec2326c6ea557aeeac3825604e/pycryptodome-3.23.0-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c8987bd3307a39bc03df5c8e0e3d8be0c4c3518b7f044b0f4c15d1aa78f52575" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f9/c5/ffe6474e0c551d54cab931918127c46d70cab8f114e0c2b5a3c071c2f484/pycryptodome-3.23.0-cp37-abi3-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:aa0698f65e5b570426fc31b8162ed4603b0c2841cbb9088e2b01641e3065915b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/18/28/e199677fc15ecf43010f2463fde4c1a53015d1fe95fb03bca2890836603a/pycryptodome-3.23.0-cp37-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:53ecbafc2b55353edcebd64bf5da94a2a2cdf5090a6915bcca6eca6cc452585a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ce/ea/4fdb09f2165ce1365c9eaefef36625583371ee514db58dc9b65d3a255c4c/pycryptodome-3.23.0-cp37-abi3-musllinux_1_2_i686.whl", hash = "sha256:156df9667ad9f2ad26255926524e1c136d6664b741547deb0a86a9acf5ea631f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/22/82/6edc3fc42fe9284aead511394bac167693fb2b0e0395b28b8bedaa07ef04/pycryptodome-3.23.0-cp37-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:dea827b4d55ee390dc89b2afe5927d4308a8b538ae91d9c6f7a5090f397af1aa" },
{ url = "https://mirrors.aliyun.com/pypi/packages/59/fe/aae679b64363eb78326c7fdc9d06ec3de18bac68be4b612fc1fe8902693c/pycryptodome-3.23.0-cp37-abi3-win32.whl", hash = "sha256:507dbead45474b62b2bbe318eb1c4c8ee641077532067fec9c1aa82c31f84886" },
{ url = "https://mirrors.aliyun.com/pypi/packages/54/2f/e97a1b8294db0daaa87012c24a7bb714147c7ade7656973fd6c736b484ff/pycryptodome-3.23.0-cp37-abi3-win_amd64.whl", hash = "sha256:c75b52aacc6c0c260f204cbdd834f76edc9fb0d8e0da9fbf8352ef58202564e2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/18/3d/f9441a0d798bf2b1e645adc3265e55706aead1255ccdad3856dbdcffec14/pycryptodome-3.23.0-cp37-abi3-win_arm64.whl", hash = "sha256:11eeeb6917903876f134b56ba11abe95c0b0fd5e3330def218083c7d98bbcb3c" },
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "pycryptodomex"
version = "3.20.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/31/a4/b03a16637574312c1b54c55aedeed8a4cb7d101d44058d46a0e5706c63e1/pycryptodomex-3.20.0.tar.gz", hash = "sha256:7a710b79baddd65b806402e14766c721aee8fb83381769c27920f26476276c1e" }
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/7a/09/668b587ddaf2aa0f94ea45bca73e7c564816fd9329a05e8f7f870425981d/pycryptodomex-3.20.0-cp35-abi3-macosx_10_9_universal2.whl", hash = "sha256:59af01efb011b0e8b686ba7758d59cf4a8263f9ad35911bfe3f416cee4f5c08c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4c/c4/9b1e8fca01c4b5a0e1c6f52ba19478b2692af4694afe8c89ebbe24348604/pycryptodomex-3.20.0-cp35-abi3-macosx_10_9_x86_64.whl", hash = "sha256:82ee7696ed8eb9a82c7037f32ba9b7c59e51dda6f105b39f043b6ef293989cb3" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4e/b9/91af61ec562b87c0932122666603a37cd17f991bc05faf9123b598d1e518/pycryptodomex-3.20.0-cp35-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:91852d4480a4537d169c29a9d104dda44094c78f1f5b67bca76c29a91042b623" },
{ url = "https://mirrors.aliyun.com/pypi/packages/20/7a/3162173af8597f0399b45c6aaa4939ccae908476fdf1b3a3cc30631fc9fb/pycryptodomex-3.20.0-cp35-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:bca649483d5ed251d06daf25957f802e44e6bb6df2e8f218ae71968ff8f8edc4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1b/43/e67f7767a76db1067008127a04617165579e6a65b5c3acb230c7383ca514/pycryptodomex-3.20.0-cp35-abi3-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:6e186342cfcc3aafaad565cbd496060e5a614b441cacc3995ef0091115c1f6c5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/bb/29/fb592db3f98b1ed330561518ff4706e869045b0cf27632a4310444731aa1/pycryptodomex-3.20.0-cp35-abi3-musllinux_1_1_aarch64.whl", hash = "sha256:25cd61e846aaab76d5791d006497134602a9e451e954833018161befc3b5b9ed" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8d/ca/7f296284fad77182ad2b2c198a7ece14b04cc9e6e905b1082c015f2254d3/pycryptodomex-3.20.0-cp35-abi3-musllinux_1_1_i686.whl", hash = "sha256:9c682436c359b5ada67e882fec34689726a09c461efd75b6ea77b2403d5665b7" },
{ url = "https://mirrors.aliyun.com/pypi/packages/48/7d/0f2b09490b98cc6a902ac15dda8760c568b9c18cfe70e0ef7a16de64d53a/pycryptodomex-3.20.0-cp35-abi3-musllinux_1_1_x86_64.whl", hash = "sha256:7a7a8f33a1f1fb762ede6cc9cbab8f2a9ba13b196bfaf7bc6f0b39d2ba315a43" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b0/1c/375adb14b71ee1c8d8232904e928b3e7af5bbbca7c04e4bec94fe8e90c3d/pycryptodomex-3.20.0-cp35-abi3-win32.whl", hash = "sha256:c39778fd0548d78917b61f03c1fa8bfda6cfcf98c767decf360945fe6f97461e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b2/e8/1b92184ab7e5595bf38000587e6f8cf9556ebd1bf0a583619bee2057afbd/pycryptodomex-3.20.0-cp35-abi3-win_amd64.whl", hash = "sha256:2a47bcc478741b71273b917232f521fd5704ab4b25d301669879e7273d3586cc" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "pydantic"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "2.12.5"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "annotated-types" },
{ name = "pydantic-core" },
{ name = "typing-extensions" },
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
{ name = "typing-inspection" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/69/44/36f1a6e523abc58ae5f928898e4aca2e0ea509b5aa6f6f392a5d882be928/pydantic-2.12.5.tar.gz", hash = "sha256:4d351024c75c0f085a9febbb665ce8c0c6ec5d30e903bdb6394b7ede26aebb49" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/5a/87/b70ad306ebb6f9b585f114d0ac2137d792b48be34d732d60e597c2f8465a/pydantic-2.12.5-py3-none-any.whl", hash = "sha256:e561593fccf61e8a20fc46dfc2dfe075b8be7d0188df33f221ad1f0139180f9d" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "pydantic-core"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "2.41.5"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "typing-extensions" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/71/70/23b021c950c2addd24ec408e9ab05d59b035b39d97cdc1130e1bce647bb6/pydantic_core-2.41.5.tar.gz", hash = "sha256:08daa51ea16ad373ffd5e7606252cc32f07bc72b28284b6bc9c6df804816476e" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/87/06/8806241ff1f70d9939f9af039c6c35f2360cf16e93c2ca76f184e76b1564/pydantic_core-2.41.5-cp313-cp313-macosx_10_12_x86_64.whl", hash = "sha256:941103c9be18ac8daf7b7adca8228f8ed6bb7a1849020f643b3a14d15b1924d9" },
{ url = "https://mirrors.aliyun.com/pypi/packages/94/02/abfa0e0bda67faa65fef1c84971c7e45928e108fe24333c81f3bfe35d5f5/pydantic_core-2.41.5-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:112e305c3314f40c93998e567879e887a3160bb8689ef3d2c04b6cc62c33ac34" },
{ url = "https://mirrors.aliyun.com/pypi/packages/15/df/a4c740c0943e93e6500f9eb23f4ca7ec9bf71b19e608ae5b579678c8d02f/pydantic_core-2.41.5-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:0cbaad15cb0c90aa221d43c00e77bb33c93e8d36e0bf74760cd00e732d10a6a0" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9a/e3/6324802931ae1d123528988e0e86587c2072ac2e5394b4bc2bc34b61ff6e/pydantic_core-2.41.5-cp313-cp313-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:03ca43e12fab6023fc79d28ca6b39b05f794ad08ec2feccc59a339b02f2b3d33" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c9/d4/2230d7151d4957dd79c3044ea26346c148c98fbf0ee6ebd41056f2d62ab5/pydantic_core-2.41.5-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:dc799088c08fa04e43144b164feb0c13f9a0bc40503f8df3e9fde58a3c0c101e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e6/9f/eaac5df17a3672fef0081b6c1bb0b82b33ee89aa5cec0d7b05f52fd4a1fa/pydantic_core-2.41.5-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:97aeba56665b4c3235a0e52b2c2f5ae9cd071b8a8310ad27bddb3f7fb30e9aa2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/cf/4e/35a80cae583a37cf15604b44240e45c05e04e86f9cfd766623149297e971/pydantic_core-2.41.5-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:406bf18d345822d6c21366031003612b9c77b3e29ffdb0f612367352aab7d586" },
{ url = "https://mirrors.aliyun.com/pypi/packages/bf/e3/f6e262673c6140dd3305d144d032f7bd5f7497d3871c1428521f19f9efa2/pydantic_core-2.41.5-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:b93590ae81f7010dbe380cdeab6f515902ebcbefe0b9327cc4804d74e93ae69d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/75/c7/20bd7fc05f0c6ea2056a4565c6f36f8968c0924f19b7d97bbfea55780e73/pydantic_core-2.41.5-cp313-cp313-musllinux_1_1_aarch64.whl", hash = "sha256:01a3d0ab748ee531f4ea6c3e48ad9dac84ddba4b0d82291f87248f2f9de8d740" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3a/8d/34318ef985c45196e004bc46c6eab2eda437e744c124ef0dbe1ff2c9d06b/pydantic_core-2.41.5-cp313-cp313-musllinux_1_1_armv7l.whl", hash = "sha256:6561e94ba9dacc9c61bce40e2d6bdc3bfaa0259d3ff36ace3b1e6901936d2e3e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9c/59/013626bf8c78a5a5d9350d12e7697d3d4de951a75565496abd40ccd46bee/pydantic_core-2.41.5-cp313-cp313-musllinux_1_1_x86_64.whl", hash = "sha256:915c3d10f81bec3a74fbd4faebe8391013ba61e5a1a8d48c4455b923bdda7858" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1a/d9/c248c103856f807ef70c18a4f986693a46a8ffe1602e5d361485da502d20/pydantic_core-2.41.5-cp313-cp313-win32.whl", hash = "sha256:650ae77860b45cfa6e2cdafc42618ceafab3a2d9a3811fcfbd3bbf8ac3c40d36" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9e/8b/341991b158ddab181cff136acd2552c9f35bd30380422a639c0671e99a91/pydantic_core-2.41.5-cp313-cp313-win_amd64.whl", hash = "sha256:79ec52ec461e99e13791ec6508c722742ad745571f234ea6255bed38c6480f11" },
{ url = "https://mirrors.aliyun.com/pypi/packages/73/7d/f2f9db34af103bea3e09735bb40b021788a5e834c81eedb541991badf8f5/pydantic_core-2.41.5-cp313-cp313-win_arm64.whl", hash = "sha256:3f84d5c1b4ab906093bdc1ff10484838aca54ef08de4afa9de0f5f14d69639cd" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: mcp server (#7084) ### What problem does this PR solve? Add MCP support with a client example. Issue link: #4344 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-04-21 09:43:20 +08:00
[[package]]
name = "pydantic-settings"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "2.12.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: mcp server (#7084) ### What problem does this PR solve? Add MCP support with a client example. Issue link: #4344 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-04-21 09:43:20 +08:00
dependencies = [
{ name = "pydantic" },
{ name = "python-dotenv" },
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
{ name = "typing-inspection" },
Feat: mcp server (#7084) ### What problem does this PR solve? Add MCP support with a client example. Issue link: #4344 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-04-21 09:43:20 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/43/4b/ac7e0aae12027748076d72a8764ff1c9d82ca75a7a52622e67ed3f765c54/pydantic_settings-2.12.0.tar.gz", hash = "sha256:005538ef951e3c2a68e1c08b292b5f2e71490def8589d4221b95dab00dafcfd0" }
Feat: mcp server (#7084) ### What problem does this PR solve? Add MCP support with a client example. Issue link: #4344 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-04-21 09:43:20 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/c1/60/5d4751ba3f4a40a6891f24eec885f51afd78d208498268c734e256fb13c4/pydantic_settings-2.12.0-py3-none-any.whl", hash = "sha256:fddb9fd99a5b18da837b29710391e945b1e30c135477f484084ee513adb93809" },
Feat: mcp server (#7084) ### What problem does this PR solve? Add MCP support with a client example. Issue link: #4344 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-04-21 09:43:20 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "pydash"
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
version = "8.0.6"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "typing-extensions" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/75/c1/1c55272f49d761cec38ddb80be9817935b9c91ebd6a8988e10f532868d56/pydash-8.0.6.tar.gz", hash = "sha256:b2821547e9723f69cf3a986be4db64de41730be149b2641947ecd12e1e11025a" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/a5/b7/cc5e7974699db40014d58c7dd7c4ad4ffc244d36930dc9ec7d06ee67d7a9/pydash-8.0.6-py3-none-any.whl", hash = "sha256:ee70a81a5b292c007f28f03a4ee8e75c1f5d7576df5457b836ec7ab2839cc5d0" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "pydivert"
version = "2.1.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/cf/71/2da9bcf742df3ab23f75f10fedca074951dd13a84bda8dea3077f68ae9a6/pydivert-2.1.0.tar.gz", hash = "sha256:f0e150f4ff591b78e35f514e319561dadff7f24a82186a171dd4d465483de5b4" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ca/8f/86d7931c62013a5a7ebf4e1642a87d4a6050c0f570e714f61b0df1984c62/pydivert-2.1.0-py2.py3-none-any.whl", hash = "sha256:382db488e3c37c03ec9ec94e061a0b24334d78dbaeebb7d4e4d32ce4355d9da1" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "pyee"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "13.0.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "typing-extensions" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/95/03/1fd98d5841cd7964a27d729ccf2199602fe05eb7a405c1462eb7277945ed/pyee-13.0.0.tar.gz", hash = "sha256:b391e3c5a434d1f5118a25615001dbc8f669cf410ab67d04c4d4e07c55481c37" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/9b/4d/b9add7c84060d4c1906abe9a7e5359f2a60f7a9a4f67268b2766673427d8/pyee-13.0.0-py3-none-any.whl", hash = "sha256:48195a3cddb3b1515ce0695ed76036b5ccc2ef3a9f963ff9f77aec0139845498" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: GitHub connector (#12314) ### What problem does this PR solve? Feat: GitHub connector ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-30 15:09:52 +08:00
[[package]]
name = "pygithub"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "2.8.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: GitHub connector (#12314) ### What problem does this PR solve? Feat: GitHub connector ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-30 15:09:52 +08:00
dependencies = [
{ name = "pyjwt", extra = ["crypto"] },
{ name = "pynacl" },
{ name = "requests" },
{ name = "typing-extensions" },
{ name = "urllib3" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/c1/74/e560bdeffea72ecb26cff27f0fad548bbff5ecc51d6a155311ea7f9e4c4c/pygithub-2.8.1.tar.gz", hash = "sha256:341b7c78521cb07324ff670afd1baa2bf5c286f8d9fd302c1798ba594a5400c9" }
Feat: GitHub connector (#12314) ### What problem does this PR solve? Feat: GitHub connector ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-30 15:09:52 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/07/ba/7049ce39f653f6140aac4beb53a5aaf08b4407b6a3019aae394c1c5244ff/pygithub-2.8.1-py3-none-any.whl", hash = "sha256:23a0a5bca93baef082e03411bf0ce27204c32be8bfa7abc92fe4a3e132936df0" },
Feat: GitHub connector (#12314) ### What problem does this PR solve? Feat: GitHub connector ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-30 15:09:52 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "pygments"
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
version = "2.19.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/b0/77/a5b8c569bf593b0140bde72ea885a803b82086995367bf2037de0159d924/pygments-2.19.2.tar.gz", hash = "sha256:636cb2477cec7f8952536970bc533bc43743542f70392ae026374600add5b887" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/c7/21/705964c7812476f378728bdf590ca4b771ec72385c533964653c68e86bdc/pygments-2.19.2-py3-none-any.whl", hash = "sha256:86540386c03d588bb81d44bc3928634ff26449851e99741617ecb9037ee5ec0b" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: support openapi (#12521) ### What problem does this PR solve? Support OpenAPI interface description. The issue of not supporting the Swagger interface after upgrading the system framework from Flask to Quart has been resolved. Resolved https://github.com/infiniflow/ragflow/issues/5264 ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: puhaiyang <“761396462@qq.com”>
2026-01-09 17:48:20 +08:00
[[package]]
name = "pyhumps"
version = "3.8.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/c4/83/fa6f8fb7accb21f39e8f2b6a18f76f6d90626bdb0a5e5448e5cc9b8ab014/pyhumps-3.8.0.tar.gz", hash = "sha256:498026258f7ee1a8e447c2e28526c0bea9407f9a59c03260aee4bd6c04d681a3" }
Feat: support openapi (#12521) ### What problem does this PR solve? Support OpenAPI interface description. The issue of not supporting the Swagger interface after upgrading the system framework from Flask to Quart has been resolved. Resolved https://github.com/infiniflow/ragflow/issues/5264 ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: puhaiyang <“761396462@qq.com”>
2026-01-09 17:48:20 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/9e/11/a1938340ecb32d71e47ad4914843775011e6e9da59ba1229f181fef3119e/pyhumps-3.8.0-py3-none-any.whl", hash = "sha256:060e1954d9069f428232a1adda165db0b9d8dfdce1d265d36df7fbff540acfd6" },
Feat: support openapi (#12521) ### What problem does this PR solve? Support OpenAPI interface description. The issue of not supporting the Swagger interface after upgrading the system framework from Flask to Quart has been resolved. Resolved https://github.com/infiniflow/ragflow/issues/5264 ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: puhaiyang <“761396462@qq.com”>
2026-01-09 17:48:20 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "pyjwt"
version = "2.8.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/30/72/8259b2bccfe4673330cea843ab23f86858a419d8f1493f66d413a76c7e3b/PyJWT-2.8.0.tar.gz", hash = "sha256:57e28d156e3d5c10088e0c68abb90bfac3df82b40a71bd0daa20c65ccd5c23de" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/2b/4f/e04a8067c7c96c364cef7ef73906504e2f40d690811c021e1a1901473a19/PyJWT-2.8.0-py3-none-any.whl", hash = "sha256:59127c392cc44c2da5bb3192169a91f429924e17aff6534d70fdc02ab3e04320" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[package.optional-dependencies]
crypto = [
{ name = "cryptography" },
]
[[package]]
name = "pymysql"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "1.1.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f5/ae/1fe3fcd9f959efa0ebe200b8de88b5a5ce3e767e38c7ac32fb179f16a388/pymysql-1.1.2.tar.gz", hash = "sha256:4961d3e165614ae65014e361811a724e2044ad3ea3739de9903ae7c21f539f03" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/7c/4c/ad33b92b9864cbde84f259d5df035a6447f91891f5be77788e2a3892bce3/pymysql-1.1.2-py3-none-any.whl", hash = "sha256:e6b1d89711dd51f8f74b1631fe08f039e7d76cf67a42a323d3178f0f25762ed9" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: GitHub connector (#12314) ### What problem does this PR solve? Feat: GitHub connector ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-30 15:09:52 +08:00
[[package]]
name = "pynacl"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "1.6.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: GitHub connector (#12314) ### What problem does this PR solve? Feat: GitHub connector ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-30 15:09:52 +08:00
dependencies = [
{ name = "cffi", marker = "platform_python_implementation != 'PyPy'" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/d9/9a/4019b524b03a13438637b11538c82781a5eda427394380381af8f04f467a/pynacl-1.6.2.tar.gz", hash = "sha256:018494d6d696ae03c7e656e5e74cdfd8ea1326962cc401bcf018f1ed8436811c" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/be/7b/4845bbf88e94586ec47a432da4e9107e3fc3ce37eb412b1398630a37f7dd/pynacl-1.6.2-cp38-abi3-macosx_10_10_universal2.whl", hash = "sha256:c949ea47e4206af7c8f604b8278093b674f7c79ed0d4719cc836902bf4517465" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1e/b4/e927e0653ba63b02a4ca5b4d852a8d1d678afbf69b3dbf9c4d0785ac905c/pynacl-1.6.2-cp38-abi3-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:8845c0631c0be43abdd865511c41eab235e0be69c81dc66a50911594198679b0" },
{ url = "https://mirrors.aliyun.com/pypi/packages/7f/81/d60984052df5c97b1d24365bc1e30024379b42c4edcd79d2436b1b9806f2/pynacl-1.6.2-cp38-abi3-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:22de65bb9010a725b0dac248f353bb072969c94fa8d6b1f34b87d7953cf7bbe4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/68/f7/322f2f9915c4ef27d140101dd0ed26b479f7e6f5f183590fd32dfc48c4d3/pynacl-1.6.2-cp38-abi3-manylinux_2_26_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:46065496ab748469cdd999246d17e301b2c24ae2fdf739132e580a0e94c94a87" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3e/d0/f301f83ac8dbe53442c5a43f6a39016f94f754d7a9815a875b65e218a307/pynacl-1.6.2-cp38-abi3-manylinux_2_26_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:8a66d6fb6ae7661c58995f9c6435bda2b1e68b54b598a6a10247bfcdadac996c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c4/58/fc6e649762b029315325ace1a8c6be66125e42f67416d3dbd47b69563d61/pynacl-1.6.2-cp38-abi3-manylinux_2_34_aarch64.whl", hash = "sha256:26bfcd00dcf2cf160f122186af731ae30ab120c18e8375684ec2670dccd28130" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c9/a8/b917096b1accc9acd878819a49d3d84875731a41eb665f6ebc826b1af99e/pynacl-1.6.2-cp38-abi3-manylinux_2_34_x86_64.whl", hash = "sha256:c8a231e36ec2cab018c4ad4358c386e36eede0319a0c41fed24f840b1dac59f6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/85/42/fe60b5f4473e12c72f977548e4028156f4d340b884c635ec6b063fe7e9a5/pynacl-1.6.2-cp38-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:68be3a09455743ff9505491220b64440ced8973fe930f270c8e07ccfa25b1f9e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fa/f9/e40e318c604259301cc091a2a63f237d9e7b424c4851cafaea4ea7c4834e/pynacl-1.6.2-cp38-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:8b097553b380236d51ed11356c953bf8ce36a29a3e596e934ecabe76c985a577" },
{ url = "https://mirrors.aliyun.com/pypi/packages/48/47/e761c254f410c023a469284a9bc210933e18588ca87706ae93002c05114c/pynacl-1.6.2-cp38-abi3-win32.whl", hash = "sha256:5811c72b473b2f38f7e2a3dc4f8642e3a3e9b5e7317266e4ced1fba85cae41aa" },
{ url = "https://mirrors.aliyun.com/pypi/packages/41/ad/334600e8cacc7d86587fe5f565480fde569dfb487389c8e1be56ac21d8ac/pynacl-1.6.2-cp38-abi3-win_amd64.whl", hash = "sha256:62985f233210dee6548c223301b6c25440852e13d59a8b81490203c3227c5ba0" },
{ url = "https://mirrors.aliyun.com/pypi/packages/29/7d/5945b5af29534641820d3bd7b00962abbbdfee84ec7e19f0d5b3175f9a31/pynacl-1.6.2-cp38-abi3-win_arm64.whl", hash = "sha256:834a43af110f743a754448463e8fd61259cd4ab5bbedcf70f9dabad1d28a394c" },
Feat: GitHub connector (#12314) ### What problem does this PR solve? Feat: GitHub connector ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-30 15:09:52 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "pynndescent"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "0.6.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "joblib" },
{ name = "llvmlite" },
{ name = "numba" },
{ name = "scikit-learn" },
Bump infinity to v0.6.11. Requires python>=3.11 (#11814) ### What problem does this PR solve? Bump infinity to v0.6.11. Requires python>=3.11 ### Type of change - [x] Refactoring
2025-12-09 16:23:37 +08:00
{ name = "scipy" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/4a/fb/7f58c397fb31666756457ee2ac4c0289ef2daad57f4ae4be8dec12f80b03/pynndescent-0.6.0.tar.gz", hash = "sha256:7ffde0fb5b400741e055a9f7d377e3702e02250616834231f6c209e39aac24f5" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/b2/e6/94145d714402fd5ade00b5661f2d0ab981219e07f7db9bfa16786cdb9c04/pynndescent-0.6.0-py3-none-any.whl", hash = "sha256:dc8c74844e4c7f5cbd1e0cd6909da86fdc789e6ff4997336e344779c3d5538ef" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
feat: add OceanBase doc engine (#11228) ### What problem does this PR solve? Add OceanBase doc engine. Close #5350 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-20 10:00:14 +08:00
[[package]]
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
name = "pyobjc"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-accessibility", marker = "platform_release >= '20.0'" },
{ name = "pyobjc-framework-accounts", marker = "platform_release >= '12.0'" },
{ name = "pyobjc-framework-addressbook" },
{ name = "pyobjc-framework-adservices", marker = "platform_release >= '20.0'" },
{ name = "pyobjc-framework-adsupport", marker = "platform_release >= '18.0'" },
{ name = "pyobjc-framework-applescriptkit" },
{ name = "pyobjc-framework-applescriptobjc", marker = "platform_release >= '10.0'" },
{ name = "pyobjc-framework-applicationservices" },
{ name = "pyobjc-framework-apptrackingtransparency", marker = "platform_release >= '20.0'" },
{ name = "pyobjc-framework-arkit", marker = "platform_release >= '25.0'" },
{ name = "pyobjc-framework-audiovideobridging", marker = "platform_release >= '12.0'" },
{ name = "pyobjc-framework-authenticationservices", marker = "platform_release >= '19.0'" },
{ name = "pyobjc-framework-automaticassessmentconfiguration", marker = "platform_release >= '19.0'" },
{ name = "pyobjc-framework-automator" },
{ name = "pyobjc-framework-avfoundation", marker = "platform_release >= '11.0'" },
{ name = "pyobjc-framework-avkit", marker = "platform_release >= '13.0'" },
{ name = "pyobjc-framework-avrouting", marker = "platform_release >= '22.0'" },
{ name = "pyobjc-framework-backgroundassets", marker = "platform_release >= '22.0'" },
{ name = "pyobjc-framework-browserenginekit", marker = "platform_release >= '23.4'" },
{ name = "pyobjc-framework-businesschat", marker = "platform_release >= '18.0'" },
{ name = "pyobjc-framework-calendarstore", marker = "platform_release >= '9.0'" },
{ name = "pyobjc-framework-callkit", marker = "platform_release >= '20.0'" },
{ name = "pyobjc-framework-carbon" },
{ name = "pyobjc-framework-cfnetwork" },
{ name = "pyobjc-framework-cinematic", marker = "platform_release >= '23.0'" },
{ name = "pyobjc-framework-classkit", marker = "platform_release >= '20.0'" },
{ name = "pyobjc-framework-cloudkit", marker = "platform_release >= '14.0'" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-collaboration", marker = "platform_release >= '9.0'" },
{ name = "pyobjc-framework-colorsync", marker = "platform_release >= '17.0'" },
{ name = "pyobjc-framework-compositorservices", marker = "platform_release >= '25.0'" },
{ name = "pyobjc-framework-contacts", marker = "platform_release >= '15.0'" },
{ name = "pyobjc-framework-contactsui", marker = "platform_release >= '15.0'" },
{ name = "pyobjc-framework-coreaudio" },
{ name = "pyobjc-framework-coreaudiokit" },
{ name = "pyobjc-framework-corebluetooth", marker = "platform_release >= '14.0'" },
{ name = "pyobjc-framework-coredata" },
{ name = "pyobjc-framework-corehaptics", marker = "platform_release >= '19.0'" },
{ name = "pyobjc-framework-corelocation", marker = "platform_release >= '10.0'" },
{ name = "pyobjc-framework-coremedia", marker = "platform_release >= '11.0'" },
{ name = "pyobjc-framework-coremediaio", marker = "platform_release >= '11.0'" },
{ name = "pyobjc-framework-coremidi" },
{ name = "pyobjc-framework-coreml", marker = "platform_release >= '17.0'" },
{ name = "pyobjc-framework-coremotion", marker = "platform_release >= '19.0'" },
{ name = "pyobjc-framework-coreservices" },
{ name = "pyobjc-framework-corespotlight", marker = "platform_release >= '17.0'" },
{ name = "pyobjc-framework-coretext" },
{ name = "pyobjc-framework-corewlan", marker = "platform_release >= '10.0'" },
{ name = "pyobjc-framework-cryptotokenkit", marker = "platform_release >= '14.0'" },
{ name = "pyobjc-framework-datadetection", marker = "platform_release >= '21.0'" },
{ name = "pyobjc-framework-devicecheck", marker = "platform_release >= '19.0'" },
{ name = "pyobjc-framework-devicediscoveryextension", marker = "platform_release >= '24.0'" },
{ name = "pyobjc-framework-dictionaryservices", marker = "platform_release >= '9.0'" },
{ name = "pyobjc-framework-discrecording" },
{ name = "pyobjc-framework-discrecordingui" },
{ name = "pyobjc-framework-diskarbitration" },
{ name = "pyobjc-framework-dvdplayback" },
{ name = "pyobjc-framework-eventkit", marker = "platform_release >= '12.0'" },
{ name = "pyobjc-framework-exceptionhandling" },
{ name = "pyobjc-framework-executionpolicy", marker = "platform_release >= '19.0'" },
{ name = "pyobjc-framework-extensionkit", marker = "platform_release >= '22.0'" },
{ name = "pyobjc-framework-externalaccessory", marker = "platform_release >= '17.0'" },
{ name = "pyobjc-framework-fileprovider", marker = "platform_release >= '19.0'" },
{ name = "pyobjc-framework-fileproviderui", marker = "platform_release >= '19.0'" },
{ name = "pyobjc-framework-findersync", marker = "platform_release >= '14.0'" },
{ name = "pyobjc-framework-fsevents", marker = "platform_release >= '9.0'" },
{ name = "pyobjc-framework-fskit", marker = "platform_release >= '24.4'" },
{ name = "pyobjc-framework-gamecenter", marker = "platform_release >= '12.0'" },
{ name = "pyobjc-framework-gamecontroller", marker = "platform_release >= '13.0'" },
{ name = "pyobjc-framework-gamekit", marker = "platform_release >= '12.0'" },
{ name = "pyobjc-framework-gameplaykit", marker = "platform_release >= '15.0'" },
{ name = "pyobjc-framework-gamesave", marker = "platform_release >= '25.0'" },
{ name = "pyobjc-framework-healthkit", marker = "platform_release >= '22.0'" },
{ name = "pyobjc-framework-imagecapturecore", marker = "platform_release >= '10.0'" },
{ name = "pyobjc-framework-inputmethodkit", marker = "platform_release >= '9.0'" },
{ name = "pyobjc-framework-installerplugins" },
{ name = "pyobjc-framework-instantmessage", marker = "platform_release >= '9.0'" },
{ name = "pyobjc-framework-intents", marker = "platform_release >= '16.0'" },
{ name = "pyobjc-framework-intentsui", marker = "platform_release >= '21.0'" },
{ name = "pyobjc-framework-iobluetooth" },
{ name = "pyobjc-framework-iobluetoothui" },
{ name = "pyobjc-framework-iosurface", marker = "platform_release >= '10.0'" },
{ name = "pyobjc-framework-ituneslibrary", marker = "platform_release >= '10.0'" },
{ name = "pyobjc-framework-kernelmanagement", marker = "platform_release >= '20.0'" },
{ name = "pyobjc-framework-latentsemanticmapping" },
{ name = "pyobjc-framework-launchservices" },
{ name = "pyobjc-framework-libdispatch", marker = "platform_release >= '12.0'" },
{ name = "pyobjc-framework-libxpc", marker = "platform_release >= '12.0'" },
{ name = "pyobjc-framework-linkpresentation", marker = "platform_release >= '19.0'" },
{ name = "pyobjc-framework-localauthentication", marker = "platform_release >= '14.0'" },
{ name = "pyobjc-framework-localauthenticationembeddedui", marker = "platform_release >= '21.0'" },
{ name = "pyobjc-framework-mailkit", marker = "platform_release >= '21.0'" },
{ name = "pyobjc-framework-mapkit", marker = "platform_release >= '13.0'" },
{ name = "pyobjc-framework-mediaaccessibility", marker = "platform_release >= '13.0'" },
{ name = "pyobjc-framework-mediaextension", marker = "platform_release >= '24.0'" },
{ name = "pyobjc-framework-medialibrary", marker = "platform_release >= '13.0'" },
{ name = "pyobjc-framework-mediaplayer", marker = "platform_release >= '16.0'" },
{ name = "pyobjc-framework-mediatoolbox", marker = "platform_release >= '13.0'" },
{ name = "pyobjc-framework-metal", marker = "platform_release >= '15.0'" },
{ name = "pyobjc-framework-metalfx", marker = "platform_release >= '22.0'" },
{ name = "pyobjc-framework-metalkit", marker = "platform_release >= '15.0'" },
{ name = "pyobjc-framework-metalperformanceshaders", marker = "platform_release >= '17.0'" },
{ name = "pyobjc-framework-metalperformanceshadersgraph", marker = "platform_release >= '20.0'" },
{ name = "pyobjc-framework-metrickit", marker = "platform_release >= '21.0'" },
{ name = "pyobjc-framework-mlcompute", marker = "platform_release >= '20.0'" },
{ name = "pyobjc-framework-modelio", marker = "platform_release >= '15.0'" },
{ name = "pyobjc-framework-multipeerconnectivity", marker = "platform_release >= '14.0'" },
{ name = "pyobjc-framework-naturallanguage", marker = "platform_release >= '18.0'" },
{ name = "pyobjc-framework-netfs", marker = "platform_release >= '10.0'" },
{ name = "pyobjc-framework-network", marker = "platform_release >= '18.0'" },
{ name = "pyobjc-framework-networkextension", marker = "platform_release >= '15.0'" },
{ name = "pyobjc-framework-notificationcenter", marker = "platform_release >= '14.0'" },
{ name = "pyobjc-framework-opendirectory", marker = "platform_release >= '10.0'" },
{ name = "pyobjc-framework-osakit" },
{ name = "pyobjc-framework-oslog", marker = "platform_release >= '19.0'" },
{ name = "pyobjc-framework-passkit", marker = "platform_release >= '20.0'" },
{ name = "pyobjc-framework-pencilkit", marker = "platform_release >= '19.0'" },
{ name = "pyobjc-framework-phase", marker = "platform_release >= '21.0'" },
{ name = "pyobjc-framework-photos", marker = "platform_release >= '15.0'" },
{ name = "pyobjc-framework-photosui", marker = "platform_release >= '15.0'" },
{ name = "pyobjc-framework-preferencepanes" },
{ name = "pyobjc-framework-pushkit", marker = "platform_release >= '19.0'" },
{ name = "pyobjc-framework-quartz" },
{ name = "pyobjc-framework-quicklookthumbnailing", marker = "platform_release >= '19.0'" },
{ name = "pyobjc-framework-replaykit", marker = "platform_release >= '20.0'" },
{ name = "pyobjc-framework-safariservices", marker = "platform_release >= '16.0'" },
{ name = "pyobjc-framework-safetykit", marker = "platform_release >= '22.0'" },
{ name = "pyobjc-framework-scenekit", marker = "platform_release >= '11.0'" },
{ name = "pyobjc-framework-screencapturekit", marker = "platform_release >= '21.4'" },
{ name = "pyobjc-framework-screensaver" },
{ name = "pyobjc-framework-screentime", marker = "platform_release >= '20.0'" },
{ name = "pyobjc-framework-scriptingbridge", marker = "platform_release >= '9.0'" },
{ name = "pyobjc-framework-searchkit" },
{ name = "pyobjc-framework-security" },
{ name = "pyobjc-framework-securityfoundation" },
{ name = "pyobjc-framework-securityinterface" },
{ name = "pyobjc-framework-securityui", marker = "platform_release >= '24.4'" },
{ name = "pyobjc-framework-sensitivecontentanalysis", marker = "platform_release >= '23.0'" },
{ name = "pyobjc-framework-servicemanagement", marker = "platform_release >= '10.0'" },
{ name = "pyobjc-framework-sharedwithyou", marker = "platform_release >= '22.0'" },
{ name = "pyobjc-framework-sharedwithyoucore", marker = "platform_release >= '22.0'" },
{ name = "pyobjc-framework-shazamkit", marker = "platform_release >= '21.0'" },
{ name = "pyobjc-framework-social", marker = "platform_release >= '12.0'" },
{ name = "pyobjc-framework-soundanalysis", marker = "platform_release >= '19.0'" },
{ name = "pyobjc-framework-speech", marker = "platform_release >= '19.0'" },
{ name = "pyobjc-framework-spritekit", marker = "platform_release >= '13.0'" },
{ name = "pyobjc-framework-storekit", marker = "platform_release >= '11.0'" },
{ name = "pyobjc-framework-symbols", marker = "platform_release >= '23.0'" },
{ name = "pyobjc-framework-syncservices" },
{ name = "pyobjc-framework-systemconfiguration" },
{ name = "pyobjc-framework-systemextensions", marker = "platform_release >= '19.0'" },
{ name = "pyobjc-framework-threadnetwork", marker = "platform_release >= '22.0'" },
{ name = "pyobjc-framework-uniformtypeidentifiers", marker = "platform_release >= '20.0'" },
{ name = "pyobjc-framework-usernotifications", marker = "platform_release >= '18.0'" },
{ name = "pyobjc-framework-usernotificationsui", marker = "platform_release >= '20.0'" },
{ name = "pyobjc-framework-videosubscriberaccount", marker = "platform_release >= '18.0'" },
{ name = "pyobjc-framework-videotoolbox", marker = "platform_release >= '12.0'" },
{ name = "pyobjc-framework-virtualization", marker = "platform_release >= '20.0'" },
{ name = "pyobjc-framework-vision", marker = "platform_release >= '17.0'" },
{ name = "pyobjc-framework-webkit" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/00/09/68336c2ed662cecc8dc71e3bdcde2666430f3eb981cbd86661e7aa260cab/pyobjc-12.2.tar.gz", hash = "sha256:7074dcb0999e611e456b12cf94b7289cf6327094359b2c21c5ae593048b909bf" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/bf/25/f988de6a2028550fbd79f386938c4d99ac0f6b7652ad212cd8b74654924c/pyobjc-12.2-py3-none-any.whl", hash = "sha256:f3b0d4cdb7d0be242a37ff27c9f0b3ef182fe8ebdbac6ae0c40ef87539fe7d77" },
feat: add OceanBase doc engine (#11228) ### What problem does this PR solve? Add OceanBase doc engine. Close #5350 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-20 10:00:14 +08:00
]
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
[[package]]
name = "pyobjc-core"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/2a/e8/a6cc12669211e7c9b29e8f26bf2159e67c7a73555dc229018abf46d8167a/pyobjc_core-12.2.tar.gz", hash = "sha256:51d7de4cfa32f508c6a7aac31f131b12d5e196a8dcf588e6e8d7e6337224f66d" }
feat: add OceanBase doc engine (#11228) ### What problem does this PR solve? Add OceanBase doc engine. Close #5350 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-20 10:00:14 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/d6/ed/6e62d038992bc7ef9091d95ec97c3c221686fe52a993a6501e961c757613/pyobjc_core-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:9287c7c46d6ae8676b4c6c0389a8f4b5381f42ae53a47151900c08b157e5a992" },
{ url = "https://mirrors.aliyun.com/pypi/packages/bf/0b/d492110202f4d1050a5e590620ebd1e730cf89f9880a26cf18205e0f5800/pyobjc_core-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:515ecf2afe168301feb66a7230d700584ce2e4b8a0ac178e19450b8898384139" },
feat: add OceanBase doc engine (#11228) ### What problem does this PR solve? Add OceanBase doc engine. Close #5350 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-20 10:00:14 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
name = "pyobjc-framework-accessibility"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-quartz" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/76/31/ddc59bea6388a37461e919f25d2120bd93df0b30f552c606fab2804c52c3/pyobjc_framework_accessibility-12.2.tar.gz", hash = "sha256:86f82cc21db65c73c72ae93b7536381f1e12a1c89d4c554a216217dacdf60bb3" }
Security: do not use litellm 1.82.7 and 1.82.8 (#13768) ### What problem does this PR solve? See [issue](https://github.com/BerriAI/litellm/issues/24518) from Litellm. Upgraded from `1.81.15` to `1.82.6`, so RAGFlow is safe as always. ### Type of change - [x] Security Co-authored-by: Jin Hai <haijin.chn@gmail.com> Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
2026-03-25 22:39:33 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/d8/1d/d13365b48fc989b5217ef94facec719f7c7ab2878e573462cf822095cb3b/pyobjc_framework_accessibility-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:e60d910261d018770722ad0987291e94229e09aef0e919073c87a30cf677686d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/28/22/2f3ebe592655f5106a072cf1f1fe331f7c9ba06b24ac8e3dcdc9e9294d03/pyobjc_framework_accessibility-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:535a3e2ccb36e707dd72ed54f7f1772ef41526cebe6fe52acc081d45a10ede15" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
name = "pyobjc-framework-accounts"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/d3/ec/aea64ed95dda48f2c1a049c638be39822f731e79ef9fb4202dd2b3d87b86/pyobjc_framework_accounts-12.2.tar.gz", hash = "sha256:823b61e3ed964efde365e6c83b605692f72cb7f7e0782dafea674f2de8e3f8ed" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/b2/fa/61661645bdc59e6aa9c92791731d688c76b84260362849f8c768e48558ea/pyobjc_framework_accounts-12.2-py2.py3-none-any.whl", hash = "sha256:a1638b7758e6371f59e7ea9f912922530062ce9640a0f1d1e66a0702bff7f8e3" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: extract message output to file (#11251) ### What problem does this PR solve? Feat: extract message output to file ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-14 19:52:11 +08:00
[[package]]
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
name = "pyobjc-framework-addressbook"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ba/03/76bd426954723343854c2f4da82bbcd61048eccce31c00381dbfb7677c85/pyobjc_framework_addressbook-12.2.tar.gz", hash = "sha256:5f763983c1c32e70f7b742e35a025530be4bf6742ff4ac98f71ab0bfeb5cf9bd" }
Feat: extract message output to file (#11251) ### What problem does this PR solve? Feat: extract message output to file ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-14 19:52:11 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/81/bc/d3afc72e56b2e49580ada60dff250669f2b80ee1e2c59bdd3e577f32d0cd/pyobjc_framework_addressbook-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:37dbd5a712c2ff71ec80cc5a86d96c98b1d5e26eb2c7bc4d251f84eb1a3c8893" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e9/a6/9d0a9433e3b76ee9e6d728d463377d1833389450315a6122342e4d42c5c1/pyobjc_framework_addressbook-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:899b19c7c2c4a8ee7020a20022a8c64daf86dccd163da03a41bbd585565d18ed" },
Feat: extract message output to file (#11251) ### What problem does this PR solve? Feat: extract message output to file ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-14 19:52:11 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
name = "pyobjc-framework-adservices"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/04/05/fb43d284efe7e48b43abe2507f79b91a4c9ebce4f76c552fc228b647c0b9/pyobjc_framework_adservices-12.2.tar.gz", hash = "sha256:77a3cf0acdf5e83c29c26be60a63e35a31ad5076ac60188a52268fe53de7cee0" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/53/2b/230ad0f52e7f960a560afbae80dacdfc831a01a3507989e1ee41755ab793/pyobjc_framework_adservices-12.2-py2.py3-none-any.whl", hash = "sha256:7c1b1f78689f66fa724c23fd20dac56e8ea3190e868d9859d08294a58e26cb33" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
name = "pyobjc-framework-adsupport"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/9e/6d/6041af76d08da1ddae42804dfff3b5fb0fb27c2da8eb7a1a04ece4532e46/pyobjc_framework_adsupport-12.2.tar.gz", hash = "sha256:fe8e730c102a6579fe905b20e497423a080fd1f6a75eabf7356d2394467f64a4" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/29/b7/956f2d841a8709ab7cae6726dcd953a54650768f9b78cf9821fb732c6cca/pyobjc_framework_adsupport-12.2-py2.py3-none-any.whl", hash = "sha256:39ca7e3c336c32c5d9d5780eba7606f4d53034bb53bc7b55c8a5a2e430ad7c66" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
name = "pyobjc-framework-applescriptkit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/92/e8/1af0305da68a6923d269806f6f60b900e473b3ac03279a8614d71e224592/pyobjc_framework_applescriptkit-12.2.tar.gz", hash = "sha256:1e1f91966b42f902d954a570a2aab8a4e24af465833525863e30c8a5b47ee4f1" }
Security: do not use litellm 1.82.7 and 1.82.8 (#13768) ### What problem does this PR solve? See [issue](https://github.com/BerriAI/litellm/issues/24518) from Litellm. Upgraded from `1.81.15` to `1.82.6`, so RAGFlow is safe as always. ### Type of change - [x] Security Co-authored-by: Jin Hai <haijin.chn@gmail.com> Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
2026-03-25 22:39:33 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/45/2b/3abd0d61208205b5617590e6c9cc35f744a41a2bdb81b5e18eba7810b9ae/pyobjc_framework_applescriptkit-12.2-py2.py3-none-any.whl", hash = "sha256:4b2a2f02e159c3c13834c3a605c1445563592f41d57ef6cdca2bd39e6409270d" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
name = "pyobjc-framework-applescriptobjc"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/86/bc/4bcbb2fdae7a4526432619ec9a15fb7ee8ef22358be422e7b3c23c280f6f/pyobjc_framework_applescriptobjc-12.2.tar.gz", hash = "sha256:118abccae98c0af67e7c850058e4651748256eba52da382a2cd58e0e7c74418e" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/6b/a8/bb375c57509a1553e5ee51758caad3f946a98180577680b33a509f65bd9f/pyobjc_framework_applescriptobjc-12.2-py2.py3-none-any.whl", hash = "sha256:0a3019b16959dd8bef9fb581894901e89e027fa0b8b21b8515deda13a3cd9b34" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
name = "pyobjc-framework-applicationservices"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-coretext" },
{ name = "pyobjc-framework-quartz" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a9/ab/1776ac62687cb3d81e59517ca55970f26efa5a96bbf3ebcea57afcdd6b06/pyobjc_framework_applicationservices-12.2.tar.gz", hash = "sha256:4f6c4027405f709872e5b098f3cd86961bdf262fb80679a548725a02171bb0cf" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/4b/28/8c85ef2dff09fb4c6adf2161bf1d32ff81dca497863682ba46107e38bbb9/pyobjc_framework_applicationservices-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:7119c75ad2c0e21b0bd44865641944e691e80abce0d5805fa344730238b16b15" },
{ url = "https://mirrors.aliyun.com/pypi/packages/61/ee/4be28e61319055092d3a963514c2fb4daba86785d4044f073a4135273bb0/pyobjc_framework_applicationservices-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:0492c478b175005f38c887523f895382a9ed47c0810ab786c6712d3fda245832" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
name = "pyobjc-framework-apptrackingtransparency"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/c6/08/28b67c30aae6b1d870aac62ef63e7f518f3507a5c8c34378ca2cd89d2f7d/pyobjc_framework_apptrackingtransparency-12.2.tar.gz", hash = "sha256:7887b9c574c8a4edafbf8e34649a77ea91aab93dcf92b22e272ab01341d9e404" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/3a/65/1a2b950fb8c7c98ee6ab894e5466094039846ff10986ceb04d4c29ce3118/pyobjc_framework_apptrackingtransparency-12.2-py2.py3-none-any.whl", hash = "sha256:f0b51c30dc8c32882aa88e891ec13e52a0b339a3bb52ab7eed162a9640b76b4a" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Add ut test into CI (#11753) ### What problem does this PR solve? As title ### Type of change - [x] Other (please describe): --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-05 11:40:16 +08:00
[[package]]
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
name = "pyobjc-framework-arkit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Add ut test into CI (#11753) ### What problem does this PR solve? As title ### Type of change - [x] Other (please describe): --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-05 11:40:16 +08:00
dependencies = [
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
Add ut test into CI (#11753) ### What problem does this PR solve? As title ### Type of change - [x] Other (please describe): --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-05 11:40:16 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/2b/27/94cce926f3a8d95050bcfa14b1fe4e6d45ff86b93f3aa53f336a9eb3e5b6/pyobjc_framework_arkit-12.2.tar.gz", hash = "sha256:48f7241f07c03a5caad05b8998ee23210f9fd2395be69b635fdcd41d11329cf3" }
Add ut test into CI (#11753) ### What problem does this PR solve? As title ### Type of change - [x] Other (please describe): --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-05 11:40:16 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/83/cf/7fc4248af3c7ef766137ffa831dbe999efeaef48df108e62d35099aa6b5e/pyobjc_framework_arkit-12.2-py2.py3-none-any.whl", hash = "sha256:6a8065f5e49c6efddfa9250f14845b5a5fc0cada5444a87d1f96fcf72ac71f19" },
Add ut test into CI (#11753) ### What problem does this PR solve? As title ### Type of change - [x] Other (please describe): --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-05 11:40:16 +08:00
]
Update graspologic to gitee (#13362) ### What problem does this PR solve? Accelerate python module downloading ### Type of change - [x] Refactoring Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2026-03-04 17:48:47 +08:00
[[package]]
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
name = "pyobjc-framework-audiovideobridging"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Update graspologic to gitee (#13362) ### What problem does this PR solve? Accelerate python module downloading ### Type of change - [x] Refactoring Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2026-03-04 17:48:47 +08:00
dependencies = [
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
Update graspologic to gitee (#13362) ### What problem does this PR solve? Accelerate python module downloading ### Type of change - [x] Refactoring Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2026-03-04 17:48:47 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/e3/cb/4e941fbd2e199b92accbee50839a7841dee74d63d6b199c0d37e7df0e40c/pyobjc_framework_audiovideobridging-12.2.tar.gz", hash = "sha256:7019abd29ab7b232618a5ab7135670e6f1e15744167b8d4f841133893914ebf5" }
Update graspologic to gitee (#13362) ### What problem does this PR solve? Accelerate python module downloading ### Type of change - [x] Refactoring Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2026-03-04 17:48:47 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/f9/d0/146911872702f1e2e25fd3274d29e796ace042d863f076947344158ffbca/pyobjc_framework_audiovideobridging-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:7b07278a182e177f0e6417d8252df6ad6b885e17d6b7d09965d35510983bf9c6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0a/48/3f73eba1e79e0eea83fe8504f64fe62688178c51ea6ec4b1f90aabeeaf3c/pyobjc_framework_audiovideobridging-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:46e962bda54b81fa0e84bd58ccd0284ee680689891c04b869542bdcbb7c3870e" },
Update graspologic to gitee (#13362) ### What problem does this PR solve? Accelerate python module downloading ### Type of change - [x] Refactoring Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2026-03-04 17:48:47 +08:00
]
Add ut test into CI (#11753) ### What problem does this PR solve? As title ### Type of change - [x] Other (please describe): --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-05 11:40:16 +08:00
[[package]]
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
name = "pyobjc-framework-authenticationservices"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Add ut test into CI (#11753) ### What problem does this PR solve? As title ### Type of change - [x] Other (please describe): --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-05 11:40:16 +08:00
dependencies = [
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
Add ut test into CI (#11753) ### What problem does this PR solve? As title ### Type of change - [x] Other (please describe): --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-05 11:40:16 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/93/a6/973aca6e96097cf2121027b09720d6e564d381c8f024ae0ebb67939d184b/pyobjc_framework_authenticationservices-12.2.tar.gz", hash = "sha256:c99c11082d5ac6c454d729142709552b10e385d99e34d8e631bedca1886c8b78" }
Add ut test into CI (#11753) ### What problem does this PR solve? As title ### Type of change - [x] Other (please describe): --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-05 11:40:16 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/e3/ef/4cd4e3648d0187aa06de7146eb7597a2ace07fe5007dab0d99d5565c1175/pyobjc_framework_authenticationservices-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:98e3b27b9033bb21e12efe89700d0d4cc1b72a90a98ded4d19c3af7edcf8eb50" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d4/ba/b1b41b65dcd3ee357fc3716e5f241ead2da57c9a8984b6a7b8c71f160d3e/pyobjc_framework_authenticationservices-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:0a0e966a24d0c1b9a8c6c94751c05e2cd4f4a47a06ec53822aa626677f52959b" },
Add ut test into CI (#11753) ### What problem does this PR solve? As title ### Type of change - [x] Other (please describe): --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-05 11:40:16 +08:00
]
Update graspologic to gitee (#13362) ### What problem does this PR solve? Accelerate python module downloading ### Type of change - [x] Refactoring Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2026-03-04 17:48:47 +08:00
[[package]]
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
name = "pyobjc-framework-automaticassessmentconfiguration"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Update graspologic to gitee (#13362) ### What problem does this PR solve? Accelerate python module downloading ### Type of change - [x] Refactoring Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2026-03-04 17:48:47 +08:00
dependencies = [
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
Update graspologic to gitee (#13362) ### What problem does this PR solve? Accelerate python module downloading ### Type of change - [x] Refactoring Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2026-03-04 17:48:47 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/2f/6c/1bb82a3487454624071fe1ba5d75e005dfd82770d1472a1cbb78e4508316/pyobjc_framework_automaticassessmentconfiguration-12.2.tar.gz", hash = "sha256:dbd96b6085cc9da3f5115d418b2e052a5807a8964e49422c0d7bf7196499f78c" }
Update graspologic to gitee (#13362) ### What problem does this PR solve? Accelerate python module downloading ### Type of change - [x] Refactoring Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2026-03-04 17:48:47 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/cd/d9/7b56f64006c83a65b38e7b58d0a2d3d499f8633d066617c5c0e1ac338970/pyobjc_framework_automaticassessmentconfiguration-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:a045df9bf1224d30727db4af8ceabc2f227356e494e93db69b4abc058b4e4edf" },
{ url = "https://mirrors.aliyun.com/pypi/packages/98/82/1846574b7a414d29f0d81344be71a25f4d18ec5e02ad1fcf1db7c2af68c4/pyobjc_framework_automaticassessmentconfiguration-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:052e81f7a83ddb45ce5c4aeb01dc57f3cc42a5496e6117eb4174d195a2c45752" },
Update graspologic to gitee (#13362) ### What problem does this PR solve? Accelerate python module downloading ### Type of change - [x] Refactoring Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2026-03-04 17:48:47 +08:00
]
Add ut test into CI (#11753) ### What problem does this PR solve? As title ### Type of change - [x] Other (please describe): --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-05 11:40:16 +08:00
[[package]]
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
name = "pyobjc-framework-automator"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Add ut test into CI (#11753) ### What problem does this PR solve? As title ### Type of change - [x] Other (please describe): --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-05 11:40:16 +08:00
dependencies = [
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
Add ut test into CI (#11753) ### What problem does this PR solve? As title ### Type of change - [x] Other (please describe): --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-05 11:40:16 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a9/3f/65daacfdd38b6936cebae9c27f25b5bcb97e4de813488caaa066c80013a7/pyobjc_framework_automator-12.2.tar.gz", hash = "sha256:bf38231f0ad38115473e853e6419c5d0511f4fe6dd904ea380160da91cf27bea" }
Add ut test into CI (#11753) ### What problem does this PR solve? As title ### Type of change - [x] Other (please describe): --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-05 11:40:16 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/16/56/1749bf5de1f138f68bfd498b639a480e06ac9b108c5e50a68cdf58379175/pyobjc_framework_automator-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:949f2d403d4d6d753122fa8ece9f9fd417b8b4afe6b65f64879259a03ca6f59e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ba/2f/de880591ffa058c8456157f5d618d5950ab8788b7f1e8f81d62ab8afe0d7/pyobjc_framework_automator-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:038ece9097bc9441554d48dac051bf445ed5496d4ced03962c5dca2ea46cfd29" },
Add ut test into CI (#11753) ### What problem does this PR solve? As title ### Type of change - [x] Other (please describe): --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-05 11:40:16 +08:00
]
Add fallback to use 'calamine' parse engine in excel_parser.py (#9374) ### What problem does this PR solve? add fallback to `calamine` engine when parse error raised using the default `openpyxl` / `xlrd` engine. e.g. the following error can be fixed: ``` Traceback (most recent call last): File "/ragflow/deepdoc/parser/excel_parser.py", line 53, in _load_excel_to_workbook df = pd.read_excel(file_like_object) File "/ragflow/.venv/lib/python3.10/site-packages/pandas/io/excel/_base.py", line 495, in read_excel io = ExcelFile( File "/ragflow/.venv/lib/python3.10/site-packages/pandas/io/excel/_base.py", line 1567, in __init__ self._reader = self._engines[engine]( File "/ragflow/.venv/lib/python3.10/site-packages/pandas/io/excel/_xlrd.py", line 46, in __init__ super().__init__( File "/ragflow/.venv/lib/python3.10/site-packages/pandas/io/excel/_base.py", line 573, in __init__ self.book = self.load_workbook(self.handles.handle, engine_kwargs) File "/ragflow/.venv/lib/python3.10/site-packages/pandas/io/excel/_xlrd.py", line 63, in load_workbook return open_workbook(file_contents=data, **engine_kwargs) File "/ragflow/.venv/lib/python3.10/site-packages/xlrd/__init__.py", line 172, in open_workbook bk = open_workbook_xls( File "/ragflow/.venv/lib/python3.10/site-packages/xlrd/book.py", line 68, in open_workbook_xls bk.biff2_8_load( File "/ragflow/.venv/lib/python3.10/site-packages/xlrd/book.py", line 641, in biff2_8_load cd.locate_named_stream(UNICODE_LITERAL(qname)) File "/ragflow/.venv/lib/python3.10/site-packages/xlrd/compdoc.py", line 398, in locate_named_stream result = self._locate_stream( File "/ragflow/.venv/lib/python3.10/site-packages/xlrd/compdoc.py", line 429, in _locate_stream raise CompDocError("%s corruption: seen[%d] == %d" % (qname, s, self.seen[s])) xlrd.compdoc.CompDocError: Workbook corruption: seen[2] == 4 ``` ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-08-12 12:41:33 +08:00
[[package]]
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
name = "pyobjc-framework-avfoundation"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-coreaudio" },
{ name = "pyobjc-framework-coremedia" },
{ name = "pyobjc-framework-quartz" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f5/d4/88d226694f8e598138fbc26808bb2d3a0637500691e42d15c64f3a8c1258/pyobjc_framework_avfoundation-12.2.tar.gz", hash = "sha256:d06443acf26f3f3e16e8a9bfad42db900af3b8c3ca34c5676343d2f8e001f56b" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/42/f5/54bcdd1543663f7e90560adc2ea4daf2ed6bd7ab92f734816cec79cd2244/pyobjc_framework_avfoundation-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:67320d3c7cb0fe1caf019d29eea38e39ede81458e3f4c13c5483892dc66f3b3a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c9/ef/f69dc23071b4b7f7a996dbeafee8afbd9be21fe2b651de2eab5d513d6622/pyobjc_framework_avfoundation-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:4b45994d9230d2751590928d0421fb7d0b7e8988e52ec1d9ab3f1d57cb869a6b" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-avkit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-quartz" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/de/5e/baee189b339f0249b35da67ff07004412abd4fbf74fbcb63b0208534991b/pyobjc_framework_avkit-12.2.tar.gz", hash = "sha256:e93667aa0d1283e692c6709b79212438e5fc7eb621371f573a7eccf3ecaa57ee" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ec/58/d1737715102c83c82789fd8f68c323ef4571de79d4bc619800287af5b6a4/pyobjc_framework_avkit-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:a9867e4eb5d84c88f962118cb22a74c594909dedd3d45eea85ead9450c397e76" },
{ url = "https://mirrors.aliyun.com/pypi/packages/53/f3/f5a8f8d154b023930534a77b2039729968d0ade2136fc1c1672b40f3500a/pyobjc_framework_avkit-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:ad719d907340151ff5c3d9b067e8c152a8a23c6d18fc8256fdbc1431b40cf7d6" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-avrouting"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/55/5f/dcb7d6a4615c0f4b80492b0ba778f0b49aba6c6b4c3a727d0bf38696aa28/pyobjc_framework_avrouting-12.2.tar.gz", hash = "sha256:78affe0c4335ff47f186fed21965f8a381e46429c1a52794044cf2835b48dd4f" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/43/51/2a34dab903467a34377de5ddc111b33506eb297fa1103a433e00a4352c80/pyobjc_framework_avrouting-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:72a6408801282037ded8a852c5d266d0e99b637496ef3aee1ac179cdd89b8d76" },
{ url = "https://mirrors.aliyun.com/pypi/packages/96/28/015dca6d3acabb21f3e93942aff5cb4420f82ded98e2761d13d932837f35/pyobjc_framework_avrouting-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:feee657352beaa05d98a00d5e744d1d7a95572baa2df34f55496ee70bfc74d35" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-backgroundassets"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/cf/3c/b344ae79feea1d4169c2e95684d9cd8b938391371b3db8f24abb98501c12/pyobjc_framework_backgroundassets-12.2.tar.gz", hash = "sha256:6df332f129cc025843f5ba6affe651bbf6499b6be35d781fbb12b52250adba9e" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/02/a9/b68ebcab2895a659cc18dcb9890aff8fb2f19f9f4e97dee2197de5608671/pyobjc_framework_backgroundassets-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:9f40f72349412f7fe9b55e083de77f1ade5c33de8be529633727c83689421a21" },
{ url = "https://mirrors.aliyun.com/pypi/packages/68/31/5e8d83b4a2269a6b8ba98a342852c5fcd6715390396c5a088e9efa88d9f3/pyobjc_framework_backgroundassets-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:1cdb5e2c45786a6a36dea60fddc90251c6482bb6d1c05b499258ecd2a6972d05" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-browserenginekit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-coreaudio" },
{ name = "pyobjc-framework-coremedia" },
{ name = "pyobjc-framework-quartz" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/22/57/b5529f648215f9abf7f7bb5f57737b75d2cb12936a28907258d6ce2846e8/pyobjc_framework_browserenginekit-12.2.tar.gz", hash = "sha256:f38924bc5e8ab80d1248b00554015b4a86ec3c3c17c779a3e43a0179671e4d69" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/8b/c4/34535b9a061dc619a2f3910d15fde604210187cda07bba4c720fedc77d26/pyobjc_framework_browserenginekit-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:e500aa3ad0c9ef5787bc72a42474d58f8b3355d16f8944dabbe5db2f15a4634d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/63/05/c0c40325e75fbcf70372b7bea10eb875b24ced3582f580205d5a58ef72e4/pyobjc_framework_browserenginekit-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:88b2b6d5cf7b1e91029e3f32fa950bbed2cf43228ef6a0869bde915caf841152" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-businesschat"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/8b/1a/9ad0a1f52f5c670948a71eceb295b6273fa7b3422651f660f9eaac5c3078/pyobjc_framework_businesschat-12.2.tar.gz", hash = "sha256:77809904d62b18377e0764458044f30ba380f462f616cb69628431814afb6f14" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/6b/19/5b9e89bf4c62bfe4ab844027f75786d171aee623fb4ca0c4dc46e67ffd7a/pyobjc_framework_businesschat-12.2-py2.py3-none-any.whl", hash = "sha256:c96bceb6796f1fb82edafcda75d5abb59e165aa0f55a6e646638ab56d94ceaa3" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-calendarstore"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/06/ae/e6ab01011033d45501d96e388eff16185152ce2459174b13e22d90c721a1/pyobjc_framework_calendarstore-12.2.tar.gz", hash = "sha256:26fd412f9c4b9bf5243d44f87b03e24c80562012f6db4ec88c78cc2ffe63e7bd" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/15/f7/356d958ef979f1dc7c01c0a94f0a0b0bf409919308b69e5eee750772f480/pyobjc_framework_calendarstore-12.2-py2.py3-none-any.whl", hash = "sha256:cc9b0cee139d0552d8b924aca13c8ea3a51caa0e2bbf57540d84f249decd3846" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-callkit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/c7/63/f050271956d5742acde26d04ec6e602af9c39e6ac338c161cdd44b083016/pyobjc_framework_callkit-12.2.tar.gz", hash = "sha256:862c4e7475cfc2ee6c663f042fc4a025cbf07edb7608aa8556e64a0591e070fe" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/87/7a/4ac154b5bef806a770de93ff5bd5cb67ac92ddb92ceef697222919a8dcfe/pyobjc_framework_callkit-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:09f8e08298a46ceb53e50c1e17c5dd9e06ac2096f1bda2f956f2ba8693e91ca6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f8/f3/72f5b9a00b7adea3cda7d911414c8deaf0bf37e0396c9c918147fa002005/pyobjc_framework_callkit-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:b7242585de40273f82c084fe00a1658f34b0357695f008c8c468ef08c50cc2d4" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-carbon"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/35/ee/91f41ac1e637b999ba80ee3d5d2ceafaef8d8709012dbb8d0a904136ad76/pyobjc_framework_carbon-12.2.tar.gz", hash = "sha256:965a291f5d5db032057e00666b5a6baed66c564451db66e005fd88be17ed8e27" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/0c/6c/4600816c385ad52f748bf527c4ed15a15d51fb79770bb0a5d6642c4f49a3/pyobjc_framework_carbon-12.2-py2.py3-none-any.whl", hash = "sha256:2420af5872473b91080b8000a50ab2e5053611dc68c094e257b75939f3fddef4" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-cfnetwork"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ce/01/3976ae7dca6f2f35659c364c2e8454a678f730f96f50d41f8ff53cb00d8d/pyobjc_framework_cfnetwork-12.2.tar.gz", hash = "sha256:d50078305b840d1c4f9d47c7c7cb66508766189f9aec6683f4207fccd6d096ee" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/07/fe/f89406f7cadd648eccb03d2777860b748f04db3245fd4bc7b1355cb00cca/pyobjc_framework_cfnetwork-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:0b4c2eba6d1060b3e0e7d63bb67d2d14b46a7e51fa8ab2c2e7767cee266f548a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2d/5d/3bb11f406df581059d65a7807dd986796aa388a33da1c816b37238c2b889/pyobjc_framework_cfnetwork-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:f4d6db3702593b2a93cf9b793f3eaac6c0fa7d128a5abcf9f8b7a65079930e5c" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-cinematic"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-avfoundation" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-coremedia" },
{ name = "pyobjc-framework-metal" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/e2/dc/afc97af264e477fea0f95b54f13350119d69d6eeeb68a6698452295576ed/pyobjc_framework_cinematic-12.2.tar.gz", hash = "sha256:2bec1954c5241e5c8dbefcab6d5b300e62f76eb3907a177840357f9de8474d7a" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ca/b6/97ff3fa5efbd7b80d40ab5be590f92b7a7ed711ce01812bd0a0a3c0453d8/pyobjc_framework_cinematic-12.2-py2.py3-none-any.whl", hash = "sha256:8df478081b8248a32e91aa2981806e05c551bc6f9ec1286b5ca2d1c64e981f6a" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-classkit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/6e/35/40a13933e42581a127334faebba2d352a01d3baab02693b31ca4622073bc/pyobjc_framework_classkit-12.2.tar.gz", hash = "sha256:41ca16e92a844bed689b634707a07ebd38bf53e1a0793bcb1005347221889d7e" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/cf/54/bb0db5a07037e4745deb8f0176cf78b03e4176bcd965693998fec86f5b4e/pyobjc_framework_classkit-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:168c875649e7e43192ef2961569fdf2ac77f662a29417991a99dff5a44dd9881" },
{ url = "https://mirrors.aliyun.com/pypi/packages/91/93/d6581548debc49ce93d94eb909d663ab1bbf15439f3febd5ba7cfda91481/pyobjc_framework_classkit-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:b593774ad4809c896d735fe3c0d69664cc74d3abe56d1a5fee809a7c05646ea6" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-cloudkit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-accounts" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-coredata" },
{ name = "pyobjc-framework-corelocation" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/19/57/418b22c326c62f03341e7a974852c5a5bae4dfce1ec88a570e22f14c662d/pyobjc_framework_cloudkit-12.2.tar.gz", hash = "sha256:7849629cc7c726c1a3e16b3324d9fac57f72b938a66f10d480f52e381bef1223" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/cc/01/1ad78a6eebbbac6381aa1163028dfe74ccc4111444ecfc1954b4d02a53b1/pyobjc_framework_cloudkit-12.2-py2.py3-none-any.whl", hash = "sha256:1e00a6d02ac005a4dd31aa0aa5e22fdd3e05f8324f91a175dfae7fd45d85afd1" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-cocoa"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/6d/cc/927169225e72bab9c9b44285656768fb75052a0bc85fdbca62740e1ca43c/pyobjc_framework_cocoa-12.2.tar.gz", hash = "sha256:20b392e2b7241caad0538dfde12143343e5dfe48f72e7df660a7548e635903dc" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/3f/3b/1af2be2bf5204bbcfc94de215d5f87d35348c9982d9b05f54ceefbc53b8f/pyobjc_framework_cocoa-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:f0bbe0abedfb24b11ff6c71e26cdefb0df001c6482f95591fad40c2688c16498" },
{ url = "https://mirrors.aliyun.com/pypi/packages/41/cb/c0435d64f1199210af36141b90aea2ae3344719f7313d4160b8b0dd527db/pyobjc_framework_cocoa-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:46b6681e2b21b099ed095339c140f2c8137d6ac5658653166ee90722f9e3c621" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-collaboration"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ed/72/f84e5835a01f42a8b14c165490e24ba72839ed411a24bd7217f4d37c7d54/pyobjc_framework_collaboration-12.2.tar.gz", hash = "sha256:77707156f677c502fa973bbb3028ad1cea8557875e75f1a03f605678c881289a" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/49/d8/a1eb9888ca7ee599bc484feb78c43453f8c154c986bdc0daff4948fe39c7/pyobjc_framework_collaboration-12.2-py2.py3-none-any.whl", hash = "sha256:219f8e5b1f0cc25dc48460ca909c09127eb80b3aa0262cb50139cd46c84b63d1" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-colorsync"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/d0/7c/eefce8ea94a1817842b54a1d6d472ae184749ca6ad8d84386436b10ac757/pyobjc_framework_colorsync-12.2.tar.gz", hash = "sha256:b403820f3504b0e728f883575637c217a2043fafca5c0b315f03c88805d37958" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/e0/c7/0562840adbfaa21c37ced637653e4ade70618e911a151974c53d25c8f5d9/pyobjc_framework_colorsync-12.2-py2.py3-none-any.whl", hash = "sha256:5d2ce9acd7ec28133facf0af900fc4a5d36971083811eb6da775be4b659f9b77" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-compositorservices"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-metal" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/c1/da/0f975d3bb2efbcd305d76f9c8869b5918116731e739cf071630ac3e1f818/pyobjc_framework_compositorservices-12.2.tar.gz", hash = "sha256:2c1e940959270aeded95b573ed57c9edd7d3da0ccd45fb5be262c93cc62e2704" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ad/86/f172e0a687dec40e7809e1efdffeb666c8d369cb73c7e50cb6f84d0b49cf/pyobjc_framework_compositorservices-12.2-py2.py3-none-any.whl", hash = "sha256:486f5e70250892122ba90f2923dc955a3622c4a047536ee3a0b15bd3568cd171" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-contacts"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/3d/fd/b1569ca626fef30123f0fa5d3cb6126e171cea0900434e4b629f15c5f403/pyobjc_framework_contacts-12.2.tar.gz", hash = "sha256:117989ffed4f41d0bbf7e81dde8154a31995d2adf1fe898c2917bc6e9c1dc122" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/b9/79/609e55416da80222dbce2ca761f70ced7e45c85244f7b66e8713eb26dae8/pyobjc_framework_contacts-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:1ee92445cb2c5168df0b25bb6d897b1389d55571eb6bba229431c1d933940a7a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/18/36/701eff1da16bb3d89dc28d0db6acf5604650dad6b722cbcba2ad9c8a6233/pyobjc_framework_contacts-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:91748bc3a42723d088f7ded819cf0155868b9bca16ed8cd753695bc71710f45d" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-contactsui"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-contacts" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/e1/e1/8f51114ec751ad746cd0aaefcb130f5dd3b09537a326425b4c8f47fc1345/pyobjc_framework_contactsui-12.2.tar.gz", hash = "sha256:ca1ab431b8838240877d8c8c1776d58c7eae9f9a7b193b0cc761162f6baf58bb" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/8c/e9/e4433cf42edaf26aaec04a4c059733b98ad9c0982a5e72da192309fbe391/pyobjc_framework_contactsui-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:8ac61436333c416fada5053c9d8b1879f82efa9bd96fbd5027da6ffcde0728b4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/db/c2/cc3c32d9811798adf6404b8204731f855714869dd0e0476b6b4d6302ef42/pyobjc_framework_contactsui-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:2c7a4f5e89a22eafc056a52caa07e96f35b82a628b42a9c47d0720634ee01099" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-coreaudio"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/c3/d0/6e4adb375b81c06fbf28d758ea60281cf78f5675a68f7f3613d5ca28aa71/pyobjc_framework_coreaudio-12.2.tar.gz", hash = "sha256:1990be9a9311869b551e6a997ee84bee12ea38fc978d5fa7fc0119caff2f9ba5" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/14/3a/d2905ed8f0bfd74d7284e4018e11bf22f027e3ce7921a3298322d2273b4b/pyobjc_framework_coreaudio-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:505564f35227b6bf3bcfcc950edeae81628724202eb2a49f4bf213be39589521" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1d/33/cf17a929f444bca49a88f4c059868eda4ecf94ab7ab89e4d47b943d5a824/pyobjc_framework_coreaudio-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:1640bf44b001c960ae7ae84a43382141711bfbee439c88dea3872c64556e0308" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-coreaudiokit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-coreaudio" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/c1/d0/189f067979cd117cfb1464dac24bd64e1d4127b09366fbf17a9bec84984b/pyobjc_framework_coreaudiokit-12.2.tar.gz", hash = "sha256:cb743400fff2cc0ee5837d124c1665135559d8505d28996423ed4a609221614c" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/f6/e4/d59c8bc9f0dc3c953887fa9b5be06ffb6b007e3fe931b4faebb533c10e1b/pyobjc_framework_coreaudiokit-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:1da5bdab7f4d3c3f6b420f3d57e28a05c620e71cd3310499d415db0bb1f65c7f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/06/69/e46f04648626f8c70cfb5e024fd52a3ab3c832e40039b7d125fe62d2a49c/pyobjc_framework_coreaudiokit-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:baaa6b2c32fe2cd2d554f0503b773710f21e061ad544773c86d7df95ddd983ab" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-corebluetooth"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/52/df/bee7ba216f9fb513710aac1701b78c97b087b37fca8ec1806f8572e0bbb3/pyobjc_framework_corebluetooth-12.2.tar.gz", hash = "sha256:8b4e5ca99953c360c391a695b0782a5328fcecafd56fdf790ad709e932feb306" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/25/eb/04f0ea541e6df3b7a328fe63ed1f508898d84910be66ae0f4d08103ad5a1/pyobjc_framework_corebluetooth-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:b43a21eb9388453c867ce82bf5d475cfce4fe2ceaea974474071f1cfd178f40f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/cd/b7/502b1e5ff1feede7fbed1f28e1fd14604cf564e4a4f5d4f33c93a4706739/pyobjc_framework_corebluetooth-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:162623f74c85041bb8a2c9a8862cee6674531d603ad7a8d65a3efc81619f84c5" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-coredata"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/dc/67/e3eb3661c72ee1bd97eab9f6fc5121987f8f8f78f47d07418c1ef6f8fd1c/pyobjc_framework_coredata-12.2.tar.gz", hash = "sha256:1942049d656e264f47177c2fbd84314839bf55f61a825ae49f4fc7b0ebb078da" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/26/44/0cec6323ea515dd5a8b5382f78e871f905aa295858716b53d4db8162d645/pyobjc_framework_coredata-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:36c1fa33a8221339983eef98c2bc9894813a51f650c30e92fbd3cd1e9f51921d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3d/a7/844e5e193cd8a5ac4620f40fb40d1d039282e17cdc19e1f50addf36e415a/pyobjc_framework_coredata-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:4ed76ef3c44a0f7f488ac445fe06878e90fedd55c6077841efbded1b125f33ea" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-corehaptics"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/38/c0/fa97c71176260bf0fb764809034e2ac4a6f3e47701d7526ff5844da8a1cf/pyobjc_framework_corehaptics-12.2.tar.gz", hash = "sha256:d55b39ea4d998dd50f980e8c1800df2558099b96a99f25813bed702ebc2bf43e" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/59/19/b4842ca6182d754360289384aa3b4adf3775a6c4ec30818c8b8c1c473792/pyobjc_framework_corehaptics-12.2-py2.py3-none-any.whl", hash = "sha256:c04c64212e7e7a0859b23b939a4375a349176317cb2b11553d388965d9a8fec8" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-corelocation"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/b1/81/f074ac32e9af964ba873ac646ceac9d372057ff8d1db01d31e8f5586a1f6/pyobjc_framework_corelocation-12.2.tar.gz", hash = "sha256:1e0f389140707ece2bb4d08def421e25b792dcc2a34000dfb128dec2e685d725" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/eb/af/a8cf57caa11c1ce96bb19855ff02fd17a7e7f1d1ba3a890369be1f731f72/pyobjc_framework_corelocation-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:8b355dcc544a2ab525744b777431cb902c951fbbf2281350b86323eb64c4b409" },
{ url = "https://mirrors.aliyun.com/pypi/packages/11/f8/4d1aa9f9736efa013cb73022df9f58beca9639d922613be0f2be16e0c709/pyobjc_framework_corelocation-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:7515eda941b03c4e4f1e6d596383d4788997a05e30ce210ae250326ca418f240" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-coremedia"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/6e/63/9b25f9abb14e8faee219e72953c3cc4bed48b0de4b86f767f073fd2fdc35/pyobjc_framework_coremedia-12.2.tar.gz", hash = "sha256:95ed2063cc48d73f6eb06f4e603372515b5b009bd2d6e547f90bbe9e64217206" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/d5/00/412b4b98b6c6b266a73be2862453d98a6f7f95e911c216ad2117831d2b18/pyobjc_framework_coremedia-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:e992147e7b0c8c435f1bb786c8c610511453cf54bdf6c6b4a17e05936bacc44b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2e/61/e3285f4b7a889a334ba06c6daef78f32e39de40ca7b061443e39e0faf42a/pyobjc_framework_coremedia-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:d9586fb8ae22ab1a500e812e5c16dde09b3ea75613a8b5ed61d17cd9ba75c809" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-coremediaio"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/40/24/7c8b41310cb874bd308c875252ec6872198f19f3be27aacdafcd98a4b27e/pyobjc_framework_coremediaio-12.2.tar.gz", hash = "sha256:38f656ebe897262463277d4551fc6cca4ac766629476576f040b34e84c956506" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/bd/4b/65ed6a4af11af67ee64b5fafbf12ef2d56dd38020e88d9b67aed2128703c/pyobjc_framework_coremediaio-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:95cfaa7b8a0c970cfacdc620991345a3b55ba641a8091a1467a94310fc11b279" },
{ url = "https://mirrors.aliyun.com/pypi/packages/51/6a/4b04551647534735d788919a8a62b5261b249654b6e427f27d12f7e5377f/pyobjc_framework_coremediaio-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:f03e5f735b99508706f9ccf5233eaae4c53399459b6c87670112dcf1c0cd791c" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-coremidi"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a9/ca/ccc156899161481cc2c22529645d590a6f570c0583567ab6ba4a0df71bbf/pyobjc_framework_coremidi-12.2.tar.gz", hash = "sha256:a00e1d906f1a1fca302c867406ed8e1e23651b1551ce1943e9495e7665b6b0d3" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/8f/a5/b87175347becf46c8ade3593da8fc660d46a5fbf2794f5d7fcb0ca4694b6/pyobjc_framework_coremidi-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:86c7114af6e06650a84eeb781b90bfb3569304f36a40e38eb2e6113e72385444" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fb/d8/2953bc296baaed3a4207131d9b90797b6a12ebee204c92addc14b6c7ac8b/pyobjc_framework_coremidi-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:ae8c2e5113fe8a4afa4ce9c5231a5bf8a2c25dccc909b80d0fdd9ca0996e6d1d" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-coreml"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/e5/a5/ee39e19017348fcba998de71f4f425abf0563b693fcf5bf7f4b84232d538/pyobjc_framework_coreml-12.2.tar.gz", hash = "sha256:884f1f68c7a74c56d26858051da8a1e19848be97f2c4b06b32ae93e505109583" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/b3/42/93b24114d5334d7a1a5031788947f14f487c0aec4529ccfd7f78611c69b9/pyobjc_framework_coreml-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:64b66797b9cc1b20d726476f78ee2e69540fad60676d5dbb120530fbd429d3c7" },
{ url = "https://mirrors.aliyun.com/pypi/packages/08/da/ccec3d91d7a740156d720056002befda423f3960691e4290460910d352e3/pyobjc_framework_coreml-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:774a28259888c1bc247e8ed5ad14d7fc8baa975591c89669cd952e1756cd7d48" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-coremotion"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/9d/38/90010239d5d6721d8670cbb0edb71c39a1abca0cf22cecbeba7cbd490de5/pyobjc_framework_coremotion-12.2.tar.gz", hash = "sha256:8a2a06809b78a72cad52f27aa830c883e19ca36b57f8418ccfaf29773c5dc5ac" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/c7/62/606ca7aa90686f1eeb48d9870409bb1132ade70afc185816f12f0bce8f30/pyobjc_framework_coremotion-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:554bd583a6c36c659e83102832b65e6829c9f4c5d7dfe30f76e4fc62ab2ae13a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f6/7b/b5a942442e0b716319c442dd7a8cc2232b682509034c832898bbb7ba3783/pyobjc_framework_coremotion-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:657fc69c6446878c0a6b5cc53f689f110a8c4bbf0d57ae87f1d19ebf7bfc2889" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-coreservices"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-fsevents" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a2/3f/92b7df81848039fa9b3eaa0181d9969f42dd8a208e8ba8b8d4a648c2cd46/pyobjc_framework_coreservices-12.2.tar.gz", hash = "sha256:ea52c77e9a162b6823c3b364f391b4369d967365e5207a0d45cf6d0e37cec0e6" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/8e/ae/ca8c1386df09bcd7175f5fa8c37cf09f458b72549843e5454b2562db7194/pyobjc_framework_coreservices-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:c07effb4e1bff841eec4304b759fe3ef73301a7b2ae3e5e3e203efe6c45e2bde" },
{ url = "https://mirrors.aliyun.com/pypi/packages/58/5a/1413029dd792fec9e9c27c18c6a48fccaad3ec675698fc2db6eb46d5e51b/pyobjc_framework_coreservices-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:fb713264a7151a92bb11e6125da06af343037b96f6f8021e7cc8f8ff42897744" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-corespotlight"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/e2/c7/ea47b22b51e9e0bc69dc4f70bf19d18c9282df5d1c11343138ee01f7022f/pyobjc_framework_corespotlight-12.2.tar.gz", hash = "sha256:6aefcef43d3d3bf3e7cbcffca053a7dad07c434aa11b117dbb39d5cf596ee829" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/7c/bf/79508c58d3f68ce8a564146dfa29ae661d9970f64f2f8615b6cba0b866a8/pyobjc_framework_corespotlight-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:d35028c872a89c0450e067c021d64b29c013fe980617aa16e9d9778f566e167b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/94/46/15463f342139c8c9d95efa0aa434494c3782472fad1ae6e0436f624eb34b/pyobjc_framework_corespotlight-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:c0d6b74f497e835f9de97aeacdab7a2c9a1544135dee1c9665b2074de44424a4" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-coretext"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-quartz" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/1f/b0/e7ef99240f853d4dddde82c9c0114cc525de7355661b2bf2d5e04cfb1582/pyobjc_framework_coretext-12.2.tar.gz", hash = "sha256:82def2c281347e0677866315675124c84c36e9bc21651d62870cfdcecb7da34e" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/cc/db/783ae8290da2edb57b479fc474c7d66a319f4fd5f1f32dd642af1fd962ec/pyobjc_framework_coretext-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:09daa6b6befea7c0d673a037d02ed13bb4443ed65d8948329ba6c8a08e06c763" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0c/58/6d0a37fed6eee8ed4c950f71cc98355e8ce8d3a38c19aba1bf7ff6ac5441/pyobjc_framework_coretext-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:867e6f56c1c7703f27a7328d42d37cd184151657a3026cf46c0de207cc90a46c" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-corewlan"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/de/82/0393b32be2551bbaf09211eff01f92c60f9c3f964d6ace23f14847d88550/pyobjc_framework_corewlan-12.2.tar.gz", hash = "sha256:e0d81a95f6923c3dea9b0ab34fb0b028f5838930a602b5aa4bf9546eb268be9d" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/d5/31/95f9eea9b71b724a239ca7445f666ec29cef9df7cc9ae3cfa77e1b9c875b/pyobjc_framework_corewlan-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:6b922017499d6921a760bf3d359b1705ef31b87a2c4cfe763452759f9dc5a91c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0a/07/461ecada55a3d164f095e6e103d1bf9011e00334d2e7ce973048a0050512/pyobjc_framework_corewlan-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:3435e7c086ae69522354ec4fb74bcb3bbaefbc4b895680dffe6fef634430d533" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-cryptotokenkit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/7e/73/f77f0dca70fee8a1999700dc878eea1e558d7d2e5088d3bce59e425f23e5/pyobjc_framework_cryptotokenkit-12.2.tar.gz", hash = "sha256:bf070e95f82db3bf8dcb0d67d459df3810bd9dfec63a31982adefba146d9e777" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/c5/ad/deaada5bb4b9f1463d3cc8b2be809818a911edf3f6618053cf6ae5307e09/pyobjc_framework_cryptotokenkit-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:240f79011ed7750a441f9d07e971ae262551bae3e6a455f7dc2dc8641fbde6c7" },
{ url = "https://mirrors.aliyun.com/pypi/packages/6a/c4/36dff9b7c73eccedc57ec8bf60948b9410a95eb27d7e94bfd4bd89586ba5/pyobjc_framework_cryptotokenkit-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:eeaa4be6eb2cc48a73a2444d5db1ad9f7c383e23d15efab12272cc66c3fc11c3" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-datadetection"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/22/fa/eb20c36de2d038e2c65629b600dc0642727ae68e96f3bc9a692cf0b7184f/pyobjc_framework_datadetection-12.2.tar.gz", hash = "sha256:5fe732751662f119c8435c608785e86edff1b2ee394fbb9f010b5613bb6703d2" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/e0/8c/080a4c6ecffe45a64db10912ecc8c777f4368f4c5ad859a584a73e7243e9/pyobjc_framework_datadetection-12.2-py2.py3-none-any.whl", hash = "sha256:23246bea4dce73c702e176ed2be1ba26afc48a627c713a91417716c3d5915ec0" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-devicecheck"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/6e/ed/0c173fbee96920addd4f2d9ad0e69dfc22b8080c16f52f6d75acd459692c/pyobjc_framework_devicecheck-12.2.tar.gz", hash = "sha256:2750bdfc7fa11e02c88fc8d681dcf202ce2f52c4eaafc55f0afa38df1bc66369" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/54/58/8dbb51f3f462ce1d1a880f8fafc47980f56e0ad1b7472e2a500c81d6332f/pyobjc_framework_devicecheck-12.2-py2.py3-none-any.whl", hash = "sha256:77bbc267426dbdf80799d9a63ed17545c2c8332e765da8f5ce34f40be78e7776" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-devicediscoveryextension"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/33/aa/848457f81670d784c3a0b35887f7d16fd76bad805ec06be21f9c15012b9a/pyobjc_framework_devicediscoveryextension-12.2.tar.gz", hash = "sha256:2f3f21f9f2c0e0f714bd38076d7082c923650ce180a15aab77f271575b4eef04" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/97/9a/eee0a7ed475068d2bc24f2fe6a8243da384a3d5c742f6b0c436322e0b592/pyobjc_framework_devicediscoveryextension-12.2-py2.py3-none-any.whl", hash = "sha256:d0f2187013cefb9dc3ff20abd8dd0260a257a6bd6cd661b4d5a364c877ee4ac7" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-dictionaryservices"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-coreservices" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f0/f1/07b79048947dcd0ed0aa2582591a17990252a5649be725c094662ef9fbe5/pyobjc_framework_dictionaryservices-12.2.tar.gz", hash = "sha256:51e826c92fc04a0aa531bd04f092ca1edb6c470bce3f04c4eff84fbedb61d929" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/3e/af/2b7cb4c630a0f9f0588a53a290d96898bb1fc8dcb2abf311c20e250d4924/pyobjc_framework_dictionaryservices-12.2-py2.py3-none-any.whl", hash = "sha256:9257ed93dad0b4bcf82c871bc587f994d7619e5916df3662cc2510fe8f85ed5c" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-discrecording"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ea/ae/ea8b74b528cb281555bea842abbaeed99db66e8270d9eec37a3b2f5a9bfa/pyobjc_framework_discrecording-12.2.tar.gz", hash = "sha256:0b1df1d250b892d1b48e1ce87c9c89e944c28a4bb708d43848f12ff2d69cc396" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/85/18/da0b7c2bf4d4cc9f6a3863ef635d72206cd6c3959ac27471db0c6b5969d8/pyobjc_framework_discrecording-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:23e4012a55a51e5af9d5a9526697efc1a0d55a1234541c58a27b464c2fdd4e2e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2c/24/e9abce96b6f19979e9701dd1dada13ac0e344425e144ee2e684ae5e62bcd/pyobjc_framework_discrecording-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:067095747927b85baf7395d16c033be15db14a61c3fa8f6d2d6ce1b06ab36517" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-discrecordingui"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-discrecording" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ad/a6/3f94dfd9508f98aa5d152f8e56dbcdd8a0ddf2bbb9f7e180a5561772a839/pyobjc_framework_discrecordingui-12.2.tar.gz", hash = "sha256:932175c1c2df93673ffd87e4d5270bd03591548e26e39664d813b5e1b0c0a47b" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/5f/bc/3d75efba436c907629af9f17414f10bf105581e6e6803d20d0ab097a4b44/pyobjc_framework_discrecordingui-12.2-py2.py3-none-any.whl", hash = "sha256:4ae5d1ec84ff47cf041973c112024f7b3b7c6eb0d5cbd3cecc393998efa0f654" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-diskarbitration"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/4e/f8/aff13f6577bae073682c7e413ce725e51a11e2262c82956b5664621caadf/pyobjc_framework_diskarbitration-12.2.tar.gz", hash = "sha256:b5ce6750cec175066ed6dc6ccc9548201642f246130c1a82d01690747028505d" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/51/c1/93db22bbc25358cf523e7a66a76de6fdfff4719ce60e86a87d368587520d/pyobjc_framework_diskarbitration-12.2-py2.py3-none-any.whl", hash = "sha256:9952012b50f8d87849ca74c56a8b6fcd9373e8b5aa4566f628165cd4a2458a25" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-dvdplayback"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/64/87/6a2aa2a54df1b18f9543a2c64baa290be6e59d98b31b46559ea47b49b5d9/pyobjc_framework_dvdplayback-12.2.tar.gz", hash = "sha256:2f504e578a809ae8544e50d9958bc2101fbb0a35921f370c810b575b70d8096e" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/da/70/c7cbd09f8a84f1fc904206f855688c76feb02aaae1ab4efb6f1858f0ec0b/pyobjc_framework_dvdplayback-12.2-py2.py3-none-any.whl", hash = "sha256:37d0b460e0783c78c3099a653ae1a7db8158b12e4da6ca91d513ec708514baa6" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-eventkit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/fd/63/1acab761201e3ba2ddaa093470bb94373a11cee4f406587fb687e31062ab/pyobjc_framework_eventkit-12.2.tar.gz", hash = "sha256:c4e96235a1f0e43ea9699054c289369efb9d7645254a6169559c671b4ed21f86" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/90/c0/257069ea9e34c3b0dc473f6ac9c03c95029e9b3c686875ae8f1c564eb0ac/pyobjc_framework_eventkit-12.2-py2.py3-none-any.whl", hash = "sha256:b56a736182365eff268b6a8c958a663d53432bac5befd3116570d3f1e4ec8b1a" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-exceptionhandling"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/6c/77/2ecf87e8ddffb3c43e0c79bba7e1113436b45c3bc30ee76a82e94f5028ec/pyobjc_framework_exceptionhandling-12.2.tar.gz", hash = "sha256:c11ab2b122b6f2d1dc4625d163a2c5713df5ec3b0372394d3b6ed875a49397f5" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/c3/24/9ab5cba61d04445d36ef2bd52ff871056cedde22c3a2a4ff60f111d1f25b/pyobjc_framework_exceptionhandling-12.2-py2.py3-none-any.whl", hash = "sha256:14a76583bec99e18c5d0b0fd1db554d6f75b614f2912435836bc4abe6e1220c5" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-executionpolicy"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f0/e6/ad5d770d04c5cd057c20362724246d1a7407883854e065c86122bcfdc272/pyobjc_framework_executionpolicy-12.2.tar.gz", hash = "sha256:9b527b0deede8057a6482a6837ad8b3f6c3117232b8c9f97a3244991c9d0449f" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/9e/25/4b8994d6bc1b4647546a89225770e68121dad788e26fb2eba458e028d6fa/pyobjc_framework_executionpolicy-12.2-py2.py3-none-any.whl", hash = "sha256:a8b6177182c1cf316696db76de23dd40b47e41e8eebbe6fa204492055c8f1f3b" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-extensionkit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/c2/37/d8578fa098ce7c4caf70bd02cc52ebae57dba3beeea260e27f10852ab738/pyobjc_framework_extensionkit-12.2.tar.gz", hash = "sha256:eb85b5ccea05f5bc41d5c134a2d199eae2a3bc4ee12cdddb94501272a45052e7" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/4b/7f/21d173aed300819ff9e7ed9918c8817f6279c0c85687208fc5466e92efce/pyobjc_framework_extensionkit-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:e81277815e9b7ad6adb8141be1ddc794cbb92acbcb8b582f0b929100c6528ffc" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a4/cf/c75dc1b6a171743eb45ba4a28f51eeb3d348727d2b85c7af32609e18847c/pyobjc_framework_extensionkit-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:7f480965e4c3d06f352116f9362ecfaee20337f16a832c3f39ec185ebfc87553" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-externalaccessory"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ea/fd/8eaf82d13c74834007c716a13a10e6c2dfc0edaf62248c4801ad951e924f/pyobjc_framework_externalaccessory-12.2.tar.gz", hash = "sha256:0df5c2db52220753e3c7673fa3051fdf5d2890b9972dc3e140edc3dc16c7cd4d" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/bb/0b/bf64b437eb48e86078cd2e9303ff4f45f1ea43b2f0febdf2bd03f44708c3/pyobjc_framework_externalaccessory-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:c04313f4b68982d9f9f23d208f234b44d80472000b9945392459eddc44a61121" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b2/98/2588a2e1888d91f1059ad0c3d1939a213f23fb5bfe0130828808eed3393a/pyobjc_framework_externalaccessory-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:e546bb0db0e6832a40ac4978a24117eb1f93c2f419f6f26d908fd4315f04bd8e" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-fileprovider"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a3/a4/44d4f784ed0601800c80e06c53e892ccc39c2a5545bb1caa13b661fd6d83/pyobjc_framework_fileprovider-12.2.tar.gz", hash = "sha256:f9b146eab8a7f664f0e417310802bc2e31ad52b0f61951ce58e471a989ad4e71" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/57/b6/27766ae4419f64f654c7c0b1f9892a0f82baec01a298fabd45d4fef10dfa/pyobjc_framework_fileprovider-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:5718203d129eaf32dbf454b2a7a5f312f2dba108a4b49fa0536702537f93c961" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3d/48/bb7466f3a3d990868eccee96c6eb02155af2aabb08b705f90302789848c1/pyobjc_framework_fileprovider-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:78c6528c5d59881f88b2fe657fb48aa75b652b830ca18f58cabc1ece2f308ff6" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-fileproviderui"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-fileprovider" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/1b/53/ce4fc44c62f813e6572426c0f66475f5a5e4804693001185faa80aa95af7/pyobjc_framework_fileproviderui-12.2.tar.gz", hash = "sha256:520974c2966057cd47206f8792d8066531551b3222c8a5824851cc19e9bba7f0" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/d5/40/4514fc9cf9b9f20d2abe1dac66c4f94b8c20da9db7865ee9dec9dd6f3f65/pyobjc_framework_fileproviderui-12.2-py2.py3-none-any.whl", hash = "sha256:0874b16ea64d055f53d0c6ede6ba61b3dbe9d2b27a64db5c12b829391a510cb6" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-findersync"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/54/8c/dc4e185d054d422f0b24417dd0a340923f7cd317e70879dcbb740bfdd01c/pyobjc_framework_findersync-12.2.tar.gz", hash = "sha256:e8c8dbfd0c1e9a055d5778c8f65a72bfaabfb8c30ff9ef635ea25ee2d607dd75" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/98/3f/cd17c17bd1e3baa57388d80c5343bceb02df455047a3cfb91498b32849cd/pyobjc_framework_findersync-12.2-py2.py3-none-any.whl", hash = "sha256:6f7a461df88f4fe0cd64c0ad326c77bf1a8f72afb1bdf1f42c1d7e02b7340dd4" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-fsevents"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/35/cb/f3fa1c24c2f7b3fc086b5cd0d7ec8bed1d5524e2b77d6e7b5c86d05284d5/pyobjc_framework_fsevents-12.2.tar.gz", hash = "sha256:c6599daed508605916fa31b23866c20ccc12565b568f61a43b5626d9b8bb8778" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/67/61/e5376e0c3a95b93a6f1a77f68ca7a454ffe892ff6c5f301f4b83231a26a4/pyobjc_framework_fsevents-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:b4adc50338dc4f9f50f3525e2f070e2ab4b469976f9d4762e373505f82068831" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3d/14/f0fc106e63187438c783f2b8480470d43fba311355343856bda91899248c/pyobjc_framework_fsevents-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:a00440f1e79f6639ffd601bf973d6c7064d82174da8433999825c33617d9c053" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-fskit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/cc/46/9f5162cfde0340792043ab60d1a039f442a67b001f7bfc981b6837a9b07d/pyobjc_framework_fskit-12.2.tar.gz", hash = "sha256:a1a7f9678c55fa947783caa24325c91727cc3045b877c6770262790c00da3952" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/5c/ce/6bec42dade6873cf440f62d18be277b342b326bdb0ee7835b5f79b2748dc/pyobjc_framework_fskit-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:fcc8db6cf9e9b0e731e056411976c184bb3861288da4e97dae91a3e3e8063ade" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8d/ce/385a29b02bc81bb70169cc2a5d85b78d3af131a113acb2615d87f12c5f16/pyobjc_framework_fskit-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:1f3935499d00678da05b97a7baa46bdbecaef3fd61768fa5dd34bd3dd484ad55" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-gamecenter"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/18/bf/71c09142969953637a524946c3609ddf472c0b4311d134223cc9c87dbf8f/pyobjc_framework_gamecenter-12.2.tar.gz", hash = "sha256:a1207696a547011094a4b8d2865a3a9018a622647a8a0371baf86a3e5959ef4f" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/41/77/2496866b618c96b3569da87385db6593688222f6fcaf733cc702abf0309f/pyobjc_framework_gamecenter-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:c5ff47f68b0f5f138fb91f03e5cc46dd12f00d95f795ed5b88f8b99c52e54e2f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/5c/d0/6ec40754993e4f6f0366d7a13e5a19b6745b7c7d6395f2be0b3b60d5688e/pyobjc_framework_gamecenter-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:ee1890cc3e1266407d41475659b382067a08765bdbc62a0ae4eecec733491fec" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-gamecontroller"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/d1/4c/440ac9652cae2fc386702a39973008fe0fd18928f70b359728c1926f5350/pyobjc_framework_gamecontroller-12.2.tar.gz", hash = "sha256:130418999cb5e202c0775fc1587666c6b9fa1d9285341eb6cbd589ecbe758a64" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/65/54/519b19e43bb3fcb8058c9cf780d324e7b3532a1b8a317ff1831615993206/pyobjc_framework_gamecontroller-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:8a448b31594c68993da348c3f2f24c02323853905e999f930d7c5d0fa6b4468d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/da/f3/11be9f022e2cc30f1ff2509bdb8405005b707beee2a01b70798054c52cdd/pyobjc_framework_gamecontroller-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:d5a5ea2c31f179c72dfbd1564aaca52fcce742c27f91e2c7af37ba1e4777c3fa" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-gamekit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-quartz" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/c6/e3/dfc67af937bdf98b92fc0f6518bd1ee17cf84025ded15bd64acb91f0db8f/pyobjc_framework_gamekit-12.2.tar.gz", hash = "sha256:d4f4f98539db6125d5fb249b845398a0534f8962e6db390822da808b91ece009" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/7e/26/f1af94c102570afb8ed0e1461fcc752098d8ea12e88959cbc3cd25709005/pyobjc_framework_gamekit-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:06ac26311a602aa0674c21d6dbc3fb0a19e0d8cc385e2e80b689518c3737df8c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/74/da/4020429066e04b02c1e40662dd546bba30adfbbbb19a3b87a4007bfe0244/pyobjc_framework_gamekit-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:d915b5a8e8073769f0c371aedbbce31152859ab6352f7488f815bd572e1a12da" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-gameplaykit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-spritekit" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a8/72/df0abe840ebc9e62f8a04afefeb38997b7774ab2b68ceebe89a9dfb81b2d/pyobjc_framework_gameplaykit-12.2.tar.gz", hash = "sha256:27da2b7c4cdef038215246fc6c98e45e076f969ebb83d827095432a8213b57df" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/68/47/58dbd455f2d3cd18598408a45b05580387e788104148993aa79b8f4749de/pyobjc_framework_gameplaykit-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:89a695937dff8a4a43ac9d389b5f3eb64ced06410e8fbeb6e04f1976ed7c8be5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/5a/73/87d2d0f4b97942f1401fd8985e8d7b2191502ed210c945472da1e37a4d2a/pyobjc_framework_gameplaykit-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:db270f78befada566106e3557bdc7281fe2f2759cf81e664e9a45ff2ce709c72" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-gamesave"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/49/fb/aed9853de0ac26c768c1bfefd7c32443367e489d16f5b5563043019aecf7/pyobjc_framework_gamesave-12.2.tar.gz", hash = "sha256:e272a1ec2fc84f680226abc6a4e034534509cbd545cb937fbc76b3a95d7b9e44" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/e1/da/570c14f8c63cda0a128084cfc53f729e9183987b4a5582cc5ce4b3dbc6b7/pyobjc_framework_gamesave-12.2-py2.py3-none-any.whl", hash = "sha256:766d6eede6e7f9ef1c43333127ce42fdfa9438e62ab94c9533d32d41eb79233a" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-healthkit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/7c/c8/fa1a309cabfa0d7190a6dc1f6194d349cdde0105b86683780190aa4e46ac/pyobjc_framework_healthkit-12.2.tar.gz", hash = "sha256:48bda5f86fff2da976c49d27c746cd190d62eebdfa6197305c3656c64a6fbf09" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/5e/44/db40ed76d01e0a99820f56dd0792448dbc34d508d2235d9185ec0fbb3db8/pyobjc_framework_healthkit-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:489858fd2cf97747390d14bd5343448a9cf612383e96958fd70968d1d7e75d02" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8b/88/839c556d8e8e75b65961e65b78f683113cca6637e1917c9c4c0e2f050249/pyobjc_framework_healthkit-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:c8d43159c2c40641f9306407ef909b338325028846d04feaf2b26673d6b1bb4d" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-imagecapturecore"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/d0/7b/fada6744e9409643f7ec948a9b8fd5632a8f935d2c1d53e8a9f443a613bb/pyobjc_framework_imagecapturecore-12.2.tar.gz", hash = "sha256:c2931ca0d40d6ffe159149e140d4d158d3f844f19b99dae35f57c34551434eea" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/b8/39/23ab07b2cef5e26b017723f061d5941416a107679f19f5327b6576cf5df0/pyobjc_framework_imagecapturecore-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:12514630d46621acaeae0b921ee7c9c9e4a2bf81a936690b767a6b745e899812" },
{ url = "https://mirrors.aliyun.com/pypi/packages/13/d3/fed9ab9b40538711e3aba6c7b92a38c5439427510dd318923237f2280dc4/pyobjc_framework_imagecapturecore-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:56dbdee10d3d735fadfa15b3e7b9a11d3574493a4114fe9bf0eae86f55b43d2e" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-inputmethodkit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f4/8d/b9ba4cf37ef153584df3098cbc19ba668aef8469f18af53ff464eed1bd96/pyobjc_framework_inputmethodkit-12.2.tar.gz", hash = "sha256:63ab93867f25a3f54fc55e98b6b0540d0d23bf6f91a0a475ff742e4221a75917" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/49/88/e5fc011ab34970cc1c1cc56b866e2edac6d160d25aafad7e86b033826cbf/pyobjc_framework_inputmethodkit-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:160ba62082a30d76b9e4aec1a4d54c95152fe39217940fea597756ad49910dde" },
{ url = "https://mirrors.aliyun.com/pypi/packages/43/68/3c2fc27aab9cf228ede204ad7c8d1fe5f4e714aa4e478fe943dde6417258/pyobjc_framework_inputmethodkit-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:0b81fba64e92953ffc1788ec13c630a9d19447deacde34a7309c0eae6412d628" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-installerplugins"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/bf/31/4284cba4b8904d2b93738249bb888762115d06876b92d30b721dea91e014/pyobjc_framework_installerplugins-12.2.tar.gz", hash = "sha256:6de54ab966f5bb304f9f5036c41937675cfdf49c44f70a1078a87dee2520261a" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/2a/08/9b8bb5f74df7d23d83ec5ab954301c91b8239091767df064aaff5b54c7a9/pyobjc_framework_installerplugins-12.2-py2.py3-none-any.whl", hash = "sha256:358ef2faefe1b9938c0563e95551ad685c4c2097a7b8bc46dbf394765eb00674" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-instantmessage"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-quartz" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/02/20/2beaee41e71fcfe9bd3149277977aaef9510fae7d4ea3d70baecf2a978f3/pyobjc_framework_instantmessage-12.2.tar.gz", hash = "sha256:95a972697cc1c8c773d2e211815c7b4f3f11ad0aa114e46ec386d92f873d5533" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/a4/63/68356d5150c203b6b027dc93048b7e81d53c58d5d437b60dce914f861496/pyobjc_framework_instantmessage-12.2-py2.py3-none-any.whl", hash = "sha256:2fe9367f736b68557bf0c57a7da2c6a854cbf46ace9f6c5b161131309dc6b262" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-intents"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/70/a1/2544e2b1e4fc794bc09b16651e967cbe11a17bf75f6d13c34a261e49a13f/pyobjc_framework_intents-12.2.tar.gz", hash = "sha256:441acf32738c8a4fa21458467ce451ce6880d0bac71b6e63e2e6634775f603b6" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/22/4b/74dc067c2475881c4c93a2f4d6f460733a963a3f687de9f413b1280e40d0/pyobjc_framework_intents-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:1786dd54d1c570ca6be16383e653b897ad4c1b0c3df9be60ac5f8e0096d4a181" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e7/8d/d6cb87fdc7df4bc81b61cb184d5fed3f55cc2cd30eed90033f434a32716e/pyobjc_framework_intents-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:9d697dca29ce2c282dafc4db9c0a0f165802e8d056a68f809bee0521ca47ff24" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-intentsui"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-intents" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/4b/10/cdd1d80716e9970f5190517daa613a2d16bfec532b3fc9bd8f987c00a178/pyobjc_framework_intentsui-12.2.tar.gz", hash = "sha256:46292de2f2e915ddff88a060639adb175e0b3b2452a45f00e440dc04acbe0e70" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/a9/f9/10f8bafdd58f136cd373fa2b0b5bd869dd6f7576abee9135fbc51c10e549/pyobjc_framework_intentsui-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:da18833b15f8f31ecc5d4769250de6883e1e117392840175ff33d002a8196f46" },
{ url = "https://mirrors.aliyun.com/pypi/packages/15/e9/6f1eb774981b665e87c08f21b3a4695d0d97dce1c2a9e7eb2ab1956605e4/pyobjc_framework_intentsui-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:96b0b96147554a0e848a6804ad725abbb183b3d045481abf54d31d6a334a1a25" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-iobluetooth"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ae/d0/ea3057fb6977ab87c85db9e676d8ec77ce77d096680bacade1964ced17e5/pyobjc_framework_iobluetooth-12.2.tar.gz", hash = "sha256:7fd7265db0f467a471a015f5a32ee79fd8f42440888d16b4206bbb985ed7331d" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/7e/7c/6252a3f61bd674c770d37517433485653aeb5e24ff2f5df2a71b79cf9197/pyobjc_framework_iobluetooth-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:8bea83069a68476036c260e3c1f3f2a6cc919aa965ecd89c8821165daaf9886f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/df/88/4e49653cc0bf64a90b1aa10b89540cd8d6929a4e4ed79d5546b3bf6cc0f8/pyobjc_framework_iobluetooth-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:2d058ea04a6924233701cb0c9312aea71174d38a37430494e725d17eb7933ae8" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-iobluetoothui"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-iobluetooth" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/27/cd/0c14569b91e29bd0f64e2496c21b15e34e789f0ca07f5cf0149dec37ec72/pyobjc_framework_iobluetoothui-12.2.tar.gz", hash = "sha256:7902974cbc2ec50adc38dd961b03eae6a84f082812ed60a440db6f690b30b2d2" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/dd/a8/2ab7199c81c9f1541e20caa78cf44df5f05162665fcfd29f16e3a85fe822/pyobjc_framework_iobluetoothui-12.2-py2.py3-none-any.whl", hash = "sha256:60622518d2f70e82398c62edf5e4e72403ef9210312945d82f3151cb5e888ccb" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-iosurface"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ca/3d/744c44cdd66f273e1ca731343e7845a281cb14a25c40d20a307334a36e21/pyobjc_framework_iosurface-12.2.tar.gz", hash = "sha256:d0315c6ad3b5ee72d3a5c946d9e92a4cace1a96bd0526f0b9f6b8009c26b9716" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/a6/f8/35fcd4d2504d0b470dd746138f47f6a8d8077fbde6aa85914c999ff9f7d5/pyobjc_framework_iosurface-12.2-py2.py3-none-any.whl", hash = "sha256:3ccd3abe40e21028419a39dbea36f60cb7e34335ab1b81aebb7a1a2f644443c5" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-ituneslibrary"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/dd/37/308813d2e80ce9c8c4e5710e0f1e1dc479be03bb06e2f28dc9058862a756/pyobjc_framework_ituneslibrary-12.2.tar.gz", hash = "sha256:780f7e5f354dc2b0a27df4abb6538f730e8a29371a8703bc05edd6ce50466f8e" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/3b/c3/6c8fa4798c8f3d5194dc7f1d4ca061840705b7a908b6ba05b3c9449a0b22/pyobjc_framework_ituneslibrary-12.2-py2.py3-none-any.whl", hash = "sha256:9876e99dac601dc523b2f0e528fb21b027693b2b6f7d697fdf460cb819339980" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-kernelmanagement"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/00/b7/f367cc6c20fc0e9b80706284a1978fa0ee90ba61c2421d7bab867a3ce4d2/pyobjc_framework_kernelmanagement-12.2.tar.gz", hash = "sha256:01abc525c1edbacf88425a36a055e52d1b4a024299097d9a2b25c34f2df4bafc" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/2b/5e/f98eb2645e1899579eb0b30695866611db6c25cef135dfdd323a164a5d8c/pyobjc_framework_kernelmanagement-12.2-py2.py3-none-any.whl", hash = "sha256:14e789ed81eaaf3ca50557015416fdc232400b682b3756efaefe4afd061552e0" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-latentsemanticmapping"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/87/bb/2824a4eae4fad404603e5ab859654f4f5c5efb9687b27f97c58915726a44/pyobjc_framework_latentsemanticmapping-12.2.tar.gz", hash = "sha256:20dfeeb005880053dcccc03dc6a58d796dd72d7f282caa625906bebc3631ecd4" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/0d/16/85b8b7dbc58a0cd0baba9854604be687620433fe354347c0e93e05025551/pyobjc_framework_latentsemanticmapping-12.2-py2.py3-none-any.whl", hash = "sha256:1c87b1dd06626eca6188c2939f0dc1f58104ac9f8979c1dd8fc5f8c7d4d901e7" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-launchservices"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-coreservices" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/25/fd/3ecceb569730898ffe9146f3e544415b4e4ea7d3343e424a3cd17500bf5d/pyobjc_framework_launchservices-12.2.tar.gz", hash = "sha256:02c3f25311673fc26eebb256cd0e873c9883575804c98b97a5d2096c8cebcecd" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/9e/8d/04c6bb237127a6c109437af6bae244b3695deca6dae67a20d408db736a1b/pyobjc_framework_launchservices-12.2-py2.py3-none-any.whl", hash = "sha256:4a0a478dfee2c53b7f3e2168f3c0e4183621050d324e7225251706500f8f5f0e" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-libdispatch"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/1b/fe/e23be301e46c30450955cdb096f16f6a86e7609787a4b8225ec24d6fdc9d/pyobjc_framework_libdispatch-12.2.tar.gz", hash = "sha256:4a41879ef7716b73d70f2e40ff39353d686cbc59d48c93217ed362d2b2baf1ba" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/83/ac/f5e01c960287e3484f71a737e7ffd75ca1c192200de70fa342a22181895f/pyobjc_framework_libdispatch-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:265fe4edfd60de111c870932ab8397f845c314332a935aa7a27100fd8562cebc" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a2/d9/8bcb4e234aee09396d06515ed80fb1f79cee3aeb67a311aa8e81a085584b/pyobjc_framework_libdispatch-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:117ed20ca54e334f657e60b9b9a3de3741d91875fd50ee51ac6a09224912cc68" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-libxpc"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/e6/a2/443c0cad2a13ec038318d391f7ebfec2189ea6d97910c0ded4c863631603/pyobjc_framework_libxpc-12.2.tar.gz", hash = "sha256:6dcae3da5ab706762d68625a391c75a3969609e5676d4091947f5c1185d4f800" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/82/f3/9e971bd977c63575359c1e4d95a0c7ac8acba1d0094b190bd738bf64dedd/pyobjc_framework_libxpc-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:7caa888b1046a7921121cb44087746803a458002a1182105b3e64734b0ca9f40" },
{ url = "https://mirrors.aliyun.com/pypi/packages/70/e7/143c999a6dde74bff96697b36270ee6d8b105ce01d11b1a066d573a5d722/pyobjc_framework_libxpc-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:56bf8fbc45e722af533bed03889d02c1dd7f81338946575bbad3efb049e8177a" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-linkpresentation"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-quartz" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/b9/20/a55fa5ca6fbeb0a0b00515f4aa37cdb946e36fbaba8c8bef6c47774c10b4/pyobjc_framework_linkpresentation-12.2.tar.gz", hash = "sha256:bb910f692f3166d6c5fce44f501a4d64a6067bac9bc26ec9488995fae56bfe6d" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/8e/42/264d202bd16bb6d8b2820c4a0d3eb1267add1c94aef68aeb749927dfff52/pyobjc_framework_linkpresentation-12.2-py2.py3-none-any.whl", hash = "sha256:68f854b4b72fef3477f1fb6604b258207a5950164e8e278330cd4848281eafee" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-localauthentication"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-security" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/3b/bb/68ff2b154ed783d39a1752cbbb7dc9d0ce55c17097dc00fe56d9080c6349/pyobjc_framework_localauthentication-12.2.tar.gz", hash = "sha256:e1d734db5ddf35093307e213115bd122ef8712463be048eabfa4062022373e21" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/09/28/23c4a20e97d06854248625f53b30543340c17e478e9d6850cbee9ea50736/pyobjc_framework_localauthentication-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:38eab810ace8b5cd5aac593b8f0db99a9851a47e7310e0a010eb33d37043da30" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0c/05/4b988f1e0f70b404b3e6ee0b0f63e9a368efb2a17c8b900e57992c126b2c/pyobjc_framework_localauthentication-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:3b7ad8257cd372950463a4d6fe2f0b5ab14891ca756a8547c3fdb1b3236ec9ca" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-localauthenticationembeddedui"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-localauthentication" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a3/93/a9d21a00896d016924d1912f2dd8918f8aa95ac0893c2b110cadc779a16a/pyobjc_framework_localauthenticationembeddedui-12.2.tar.gz", hash = "sha256:cff4e489375b3898f34d8ab378f8d84623ab19e89a35de958825c125aafceca6" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/5d/94/2e453acd660d9e1e059269c72a95b9f1f2de4a3b7a5dc0f6d6cdfd0eed5e/pyobjc_framework_localauthenticationembeddedui-12.2-py2.py3-none-any.whl", hash = "sha256:0b306917aa011deb364e85c118624d2d80c3eaf67016a345a6c4bc4960416b11" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-mailkit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/0d/44/b8fc4dec34b6f29a98f64f38eb6f37e9729aeee4d2e90887a9ece06010eb/pyobjc_framework_mailkit-12.2.tar.gz", hash = "sha256:e29ea94210faefc46a4aefe0c1647bfb77b42cfc2a4962fb8316cb967b34c47f" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/b9/95/2b4b95d7e5e43750631596da4ee8de4800883475864ca9f169881fbefb3b/pyobjc_framework_mailkit-12.2-py2.py3-none-any.whl", hash = "sha256:25b9aa8c513c40d931a7c5cc44571fd090e3d565f2e1153a634f6980b08da733" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-mapkit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-corelocation" },
{ name = "pyobjc-framework-quartz" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/74/6b/0f3d14cd1b6f3d30420e268fe6b97f6ab55f1717a4c436ac6e57576c480f/pyobjc_framework_mapkit-12.2.tar.gz", hash = "sha256:65fdf104bd8a1d3e8965c689446b7e42a44ea4bc2da5c871ff389ef7a14ee030" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/1f/23/649ecd0f1b7092390d5012bfccf0d653840a002c215fe5c7b1d0e1ffb069/pyobjc_framework_mapkit-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:4dc30cdae3b30869488ffa936d1bba327b6454809dc1d584083a8033637ab7f3" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a6/ee/d9a3b6a982a3e7fd04d06d1da9627c76b08c7107e46fff217f35a60f0350/pyobjc_framework_mapkit-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:4245ce0af24948471eca1da30dac432712fb11ce0f65100420bfd74d176359bc" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-mediaaccessibility"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/02/63/34d6f7af511b116dca7761385b7462eda9f39641a9ead9eb1404d621e571/pyobjc_framework_mediaaccessibility-12.2.tar.gz", hash = "sha256:9036a6412bce491b8ca477d4673b6e441cbf2cb2463d617f82bed818c5bde9d7" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/3d/b9/fb53b336b9d6233ca96e53aec4ea142d49b225f846bbd9e13353477f118b/pyobjc_framework_mediaaccessibility-12.2-py2.py3-none-any.whl", hash = "sha256:2e1d023c738ef09ed57635ae277b3644a09d7313e6bb979ee1c6d68bae57e4a8" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-mediaextension"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-avfoundation" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-coremedia" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/60/db/c3c4a933685423d16f69daf8a0dbe026fd45a05e0988ed4842dc2a286097/pyobjc_framework_mediaextension-12.2.tar.gz", hash = "sha256:9affa99bd94774ed91f3fc9f314468b4dc3bc6fc110ee8a81884d3947a9ced12" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/0e/8b/1576473cd8154bc5e2b36cf2f82bac60ad1223bc09145248639446bcfd18/pyobjc_framework_mediaextension-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:37e8a452f17db3bce2b50dec8a20b6851eed88906809d2361385c3b096751ff9" },
{ url = "https://mirrors.aliyun.com/pypi/packages/7a/31/473101b145d98dc738c021d66496149d38d2b57d746cf69a3665e5403dfc/pyobjc_framework_mediaextension-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:5326b3ca9f8ba0f2b51525829bc833f9c441980e3cf6628137e51a50d9e45cbf" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-medialibrary"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-quartz" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/50/44/ec97fab19e92f5d41c4e0f333bcd54f1f76a255406eadd93a95d753da711/pyobjc_framework_medialibrary-12.2.tar.gz", hash = "sha256:9b65eb789cf10b20f2c5bc4d5f4e61e5feb02a43bc213ac3c03b647e30b96634" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/1d/ee/8583d8c8a738e13f3e70bfb84ad70aab17f2995af55ece36780c1acbfdd1/pyobjc_framework_medialibrary-12.2-py2.py3-none-any.whl", hash = "sha256:3bc97cb03e633a3f6f0a4e9d351210000100ebd1a4b19624c49c4e9bc7b5e574" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-mediaplayer"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-avfoundation" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/c0/c0/aec081f84830e483e23a1763339c9c6ef9bd868d44d9aa39a83080b0ac29/pyobjc_framework_mediaplayer-12.2.tar.gz", hash = "sha256:ee4f818a85e89a4c691e368b2709709d11658b81713fe367342af24dba2c5e62" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/1f/89/a12bf5f69920b8909cf92b3e7722082600db4262293d5b878b58eaae8f7a/pyobjc_framework_mediaplayer-12.2-py2.py3-none-any.whl", hash = "sha256:436d3b410b84c7fa6577c4774faa4acc4bd3ca79f582b183e281ca63429a1574" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-mediatoolbox"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/3f/6e/5181efc8c1a077599ab5ac1cb2263ebb43b1603ed994e486f8a983d2e694/pyobjc_framework_mediatoolbox-12.2.tar.gz", hash = "sha256:35fbaa7df491df5b756da1ad2035f0f04f72e556a560cdd9a6ef74ce315c555a" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ea/5c/3bdcef52f234ce16c9b06aabfac43caeeea68a5068425bdd510e6277c256/pyobjc_framework_mediatoolbox-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:13cca0632cbb91f5c207ec6a9001aebb4bdd1077d3753a5bc7fef7f82c3e9a57" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a2/6b/8ca72d9579f84560f5a648973032757e5b0c61bf1eb74e763bdca6913ef7/pyobjc_framework_mediatoolbox-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:5b876927d92fb35f6ca6a5c8e030cb853a8b5c68add7d59ebec4ae4358a1f5af" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-metal"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/71/a9/ff281c31ce7d0cb819af1989b77258b2d8cffc0b9fad47fcdd39043f4d81/pyobjc_framework_metal-12.2.tar.gz", hash = "sha256:4fb2cfd42cc8e808f08f1f2fe57de713734dd2f495c666296b2c8a5fba9be6dc" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/dd/90/f3832396e967ae149b480ebf31da0ab7b033c145b15274b2bd5357d8faf5/pyobjc_framework_metal-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:538826eb6020e2d49e481f7ad390319dfa1016aeeac623f8b76bfe8aaec8a66f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ea/95/f36313fab765e90c8400c9423cc0f6af6fec96a919ad894b09681d0b4a88/pyobjc_framework_metal-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:b9a9ba61ab834ada0e23fe7d560cf01e6948a88a4aef3342c032683b795bc908" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-metalfx"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-metal" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/87/06/e29f5153fce867015298a9b0b0365777004ff5f4e5276728135afc496269/pyobjc_framework_metalfx-12.2.tar.gz", hash = "sha256:46774b9c938f2af40510a787be291dbc0c7e3778494704489e728fc349183244" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/c8/10/40bd25526328be5bb7aa14b481146fee3dbb1271662b5b69112f47f2833e/pyobjc_framework_metalfx-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:8f079710065e4f4628c34a2528b0dd868c9d16e3af58230d5f18f6bd22d2ee74" },
{ url = "https://mirrors.aliyun.com/pypi/packages/46/d7/5189782d95c4790b704f70dff50951c183e0567aa53fd4892ccbc91b6f8b/pyobjc_framework_metalfx-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:40c7e259991c04f5520c4c40eea02810c50812f2bac2c5ff845828a35d99ccbe" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-metalkit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-metal" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/14/48/3bbf8a296054633a8a297549c919b3ff6f53c469bdcb0046d31fd7c92ee4/pyobjc_framework_metalkit-12.2.tar.gz", hash = "sha256:0f6379b74d64ce9cb864476caa277ab5010da3b0779b7fdc94af7da25f013a1a" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/1d/3d/982672d8b3b97e59214b85e694482f8d4b373d540f3d4eb1ad27c7763b0b/pyobjc_framework_metalkit-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:d1f1e352621d5e89a2066f992a2b03091d180fd4ee3b6c7b58b432e22d76641b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/87/24/163bbd1c430a7c6adfec59ad92c3f1ec7f3de0ab6739e2f5839edf90f8c5/pyobjc_framework_metalkit-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:a855935379ededd36ab55507af01b796b845a0acad7dddcff9d9f305fd903b82" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-metalperformanceshaders"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-metal" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ad/77/9675e6e2ea06d693a75586c4194630c4346e6661bcf9c9299353dde1edf3/pyobjc_framework_metalperformanceshaders-12.2.tar.gz", hash = "sha256:f428f762bcf552f1a8f5533508ae4c73057a7f4f4a93bc997a921edec63b51e6" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/50/f6/2eaf4c358eb9684b13a08f73239c654029805da1be0ae5052ac0e15fafc5/pyobjc_framework_metalperformanceshaders-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:ac154ee0c5bd0c715d2b837689a070b811994a772b209784907730793b725012" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e0/f4/f134c898df89217ae3d07d364c75ac67048dd9106a164d7dcda52f83e5c6/pyobjc_framework_metalperformanceshaders-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:18cda4e2ed06da0caebb11fa54596e273e8d50cdbf243aa1d34fa347da71db77" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-metalperformanceshadersgraph"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-metalperformanceshaders" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/e3/95/c274ffa29063e6f331a2d7d6942610ea2dc1e1984c647e68aff0f93e35b3/pyobjc_framework_metalperformanceshadersgraph-12.2.tar.gz", hash = "sha256:16249c5cd8d8403b5d81b871b618fdc0a8af0d7c81b461e733c0b5c1f9a9ef80" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/0b/ca/173cf282b48c454dac9ef40f96b0e5338077803f3fbf0f07c09333d3a9aa/pyobjc_framework_metalperformanceshadersgraph-12.2-py2.py3-none-any.whl", hash = "sha256:e465d7717df4b000e3a529054cdf547e50e175f321b53c9520bcd6c69c08c837" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-metrickit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/00/7e/b880dd8c641abbc0807d75396132df80d7d9cb0d64006b078c7f2e169fb8/pyobjc_framework_metrickit-12.2.tar.gz", hash = "sha256:c76cae82489813ca121de0ad6013555aa826f8742adf975028b3e4268ac51e6c" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/2e/7a/224f8a506539197efb51df3aa84b269262ee14a6a4ecf9ac08d7267dbb54/pyobjc_framework_metrickit-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:db64bd8d99011d67edb16bdb0f923191d95649ee7bc4a276d4f2516e975ec780" },
{ url = "https://mirrors.aliyun.com/pypi/packages/71/c8/a112d3aa27eb2749d1308122b40fb1b45947199034e37ee5d86759eaeff7/pyobjc_framework_metrickit-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:eb146093ab8ec597e74307ece22c572b3654a4247d6e0d0fd3f4635f0166286e" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-mlcompute"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/98/32/72a3733b6f3b1c60a18acfa517d28f9bcfff9822a1e21298c50c30c1a944/pyobjc_framework_mlcompute-12.2.tar.gz", hash = "sha256:7a2108c89ccae06e0fa03b1ef08c37bd9fae0ea0727d8b5890254bf44347eeac" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/0c/7f/44c69bb026a7ffc00686a51514b43d7fe1a09682997cabe7a39a016aad0b/pyobjc_framework_mlcompute-12.2-py2.py3-none-any.whl", hash = "sha256:18066ab867e02f5eb2cc66145b4274e6a7105e69165550356ec4a75937db1aae" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-modelio"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-quartz" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/c0/42/ff2d205133a753bd95c937b4591559bd10417d88da7debcdb7ef22e00dcc/pyobjc_framework_modelio-12.2.tar.gz", hash = "sha256:3f492868aa12c7107e456238fc718cac0316e8d95f7680b7465e41918d72db2f" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/57/92/6dc491a2ffe4c852c266d2a3430114fa0a4cee057930556adb5f27e4948c/pyobjc_framework_modelio-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:a2bcdc8079e4886dbee3c2351da5dfc53fe46d1d8e73a9a751085adce97a74ff" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f7/f6/c4253cdf4ce3d2b55465805499deeecaff9230425366bdb9266c0e811736/pyobjc_framework_modelio-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:c7999915fad52b6578688d32ab4c761460237cb28b06dd663f81993fa6c4580e" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-multipeerconnectivity"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a7/a5/3112f721c54e1a09d12d8dc48b165ff51a1147783d8d63a53e1da5b603ca/pyobjc_framework_multipeerconnectivity-12.2.tar.gz", hash = "sha256:7a5879cb83e9a7ea8c8d9e7e11ed3de4c9d67046e6c330b8a5f6332339f631a6" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/8f/50/4080a78aaa2fe925dd2473fde41f650e3433cbc658c062d09b3658e181a0/pyobjc_framework_multipeerconnectivity-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:ff0c6305c3b573d97f217a69b8b82b9075d788a82768377a07a456114bec248d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2c/75/50488445848adbac42d752d99d126913dadf5e7026aaf28d83c836b6f788/pyobjc_framework_multipeerconnectivity-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:5f5de878761b5e1cec4566df0aef13143845a83e4d12c80faeaefd308fd919c4" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-naturallanguage"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/6a/1b/8fe870dd9ffed8943e72328fdd71df69363ea55037107882a3c11182edbe/pyobjc_framework_naturallanguage-12.2.tar.gz", hash = "sha256:f6980ba957bb24501bbe135ffcdd0ff149f33d30e965f033ad153df756b33cc7" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/f5/00/5923b25edd9d64ba7494cb8e28c1c073ec5cbb43a41eaa70bffa4b7a2862/pyobjc_framework_naturallanguage-12.2-py2.py3-none-any.whl", hash = "sha256:8c02a9ea25b888eefb0ac503211e17a9868a1781f5d5bf9c1655445abc14ffe2" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-netfs"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/de/c8/e22988cf11df9b7c7da2ba04b7e9b9435fa7dc4ba3be7d6bb34d60f686eb/pyobjc_framework_netfs-12.2.tar.gz", hash = "sha256:fdb04ecc91864a4b79498a07f4d34f33b5dcae34958d555f5cbd69e20165991d" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/f8/c1/1584c42fc716ee13d085461913b59f5682fddd76b8094feb760477a8edab/pyobjc_framework_netfs-12.2-py2.py3-none-any.whl", hash = "sha256:864d09a7b671f4407ff577739949c98ae8ba9b013433cf938fb0c86319151248" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-network"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/fd/a8/12fcf0341ff377600a102f48266ef85e67af9dc7d894ca40fb18dd8702bb/pyobjc_framework_network-12.2.tar.gz", hash = "sha256:99d7a1900d8250d008eb285b20792d14076eedd665749cb0f48e1e102056207a" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/c6/b3/c99e31a29975fe58d81396105518d74cdd1b0de5f8d939e5a5892b7d95f7/pyobjc_framework_network-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:6677fac4bfe2d8ae320ffc069767c0e2ca7d50304ad3cf7a77d0a4d08d69465e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/85/94/233e77df7b4b53aa0316346540ee1345ed6c2d5b1893fd10fe268e1d9a1c/pyobjc_framework_network-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:c198226431964487f1213ba17de69eddcaf5ddd1bea406b31c6845cc956da536" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-networkextension"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/11/f0/bf65ac7714c81159f295d9386d718c9a876fc212f8af54b6da22c2d2873a/pyobjc_framework_networkextension-12.2.tar.gz", hash = "sha256:45bc3e7966cb9fce997832960d8aaf324a187f84019e65fc4432c27046715f70" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/3b/86/91d135890a83c021e2b031c467001802314a5b2c6aa1580669b1bf524060/pyobjc_framework_networkextension-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:631ca462cedf277531800e3cbdf6d4f98aac2dd2ce8512fc855076fa373d6725" },
{ url = "https://mirrors.aliyun.com/pypi/packages/85/d5/47debfb91e4f0c1bdbb1d4f433662cdc97e0467bf84379764be7dc26dfe3/pyobjc_framework_networkextension-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:1f9c7e7aa156d604a2c0764f65ef95131ff0c7724dd9ff35347d96b27836814c" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-notificationcenter"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/e2/d4/62b27947aa2ff2228b49083245e622eb4840f68c58d6d3e89041f597b5dc/pyobjc_framework_notificationcenter-12.2.tar.gz", hash = "sha256:d7a20fd67851d91db142765aa823a5c60c03c37957cea310a4eaad37640c085e" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/46/ac/2b2465db73f0340538df30728e5cc4a1c3014066651b091406cf2fe96e0e/pyobjc_framework_notificationcenter-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:d869a7689086ccb6f5c82a29684f38cbd2293e63a50fb8b06cec6151b497a73e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/11/0d/bff8fbabbad5f0328852f0d9b7d87551b206c3c76ff9c5ecd863d7613643/pyobjc_framework_notificationcenter-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:db911fd4de68f52abda8e812799e87721092f78ac575ec50bff82018e1efb06a" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-opendirectory"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/12/43/1c4b9f4e3739c3ab763f42b6e51ea74afebafefa5662b2740a6fb812697a/pyobjc_framework_opendirectory-12.2.tar.gz", hash = "sha256:e9edbe26c4d9aa533f343264d34af96fbef958d0c3a833acd2d64cf0fdb138e7" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/09/0d/6926ea3ba0f58e52e7a2260b416abd44c4680dc6c5c8e2212ccdd8d649cc/pyobjc_framework_opendirectory-12.2-py2.py3-none-any.whl", hash = "sha256:71d17cc1be29dd2ac50ae76fd654d233bed3b117854bf0ace021d8242a5a5566" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-osakit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/9d/bf/f6ea6b2e15151c79a22feda80d9e2c120d457cf16750c7751408f3dffc7c/pyobjc_framework_osakit-12.2.tar.gz", hash = "sha256:219efdbe36fa6bea1c780b28ad8a1b23ec4ca0ee9ad3cf040da192f883514b81" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/f4/ad/e79528da31d07f4d227c0f674b94a8a5b0aeac062d6a2df8363dbf0d99c9/pyobjc_framework_osakit-12.2-py2.py3-none-any.whl", hash = "sha256:604f428b00a0b1da1f40fc8a78db96e08b23abc7d666545a0909cc6e7b5ca2fd" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-oslog"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-coremedia" },
{ name = "pyobjc-framework-quartz" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/64/ba/dc92bab4a4be642226c3f6b71e0a100f70c27027728df00eeaaa588d04f8/pyobjc_framework_oslog-12.2.tar.gz", hash = "sha256:c2213dbb3edff318c731b300509c69bd922ad102311042dddcd9de8287e50a83" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/be/13/df0e2bae8e19998101ba238a2cd27c147906c63cd42d1814293bc08f729b/pyobjc_framework_oslog-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:ce52bdc6803bf980369d3f265339063fb71391fcf980acadf9f73c4b7af7413d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c5/cb/b73921c1ce8a78620c14f00f2907de7430a5dd78cd858b3b72cf43758ee1/pyobjc_framework_oslog-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:102c32f34f772befc22d37bbf3a6a70a88b3389ea6d25111217e84c5bfde40cf" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-passkit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/03/ed/68c8657d080f53706ea7e6cd77a4babb2b8f95a0c73471db5a56c5742988/pyobjc_framework_passkit-12.2.tar.gz", hash = "sha256:14a9b37b290942e86633ac59f923195d0038ff0f746443d914dfd5858cba9eff" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/3a/b0/dc2ca2ccf1377eb980220ea1b392803c0d7f42b2164960ed5a04361ec592/pyobjc_framework_passkit-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:3ca56bc17cd5ac7fea74caec55f3329a2b21c9462f70f1799a1896cb2fc9c06d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a2/24/1355ea6c97237c844e689e1f6b92481c3382693590af67a637a7c5205df7/pyobjc_framework_passkit-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:f58e668a4e827e2bfb7ddc41aae68025c980b5ec25d218ea8a49740159268f8e" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-pencilkit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ea/f0/39d09038bea6f2130ed6ad9de9034a2ebf3babd9eb5ad75a70f60160cfd6/pyobjc_framework_pencilkit-12.2.tar.gz", hash = "sha256:a224250d5bd8490e619792570bf2b03849b875757c076412ef83bddf58a3d2c5" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ee/d6/45dc7133e4d561b725718a2d8cd016605a6fe7bcf3f12a002c34a309ee4c/pyobjc_framework_pencilkit-12.2-py2.py3-none-any.whl", hash = "sha256:aec57e9ebe2a875d4fc17c46ef16ffcee74ba7551dcd6d83207efb05c3377187" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-phase"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-avfoundation" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/40/4e/3102eb1fa234c851e23213701d3cda26eb3516c61598dcc7e3f26eb17724/pyobjc_framework_phase-12.2.tar.gz", hash = "sha256:fc1ec192bf7aac2627425c5f1fdabc3df614d100d18a6273b25ec4593f257eca" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/a8/80/9d526e002b9be65dd80abc2cc5c62afe8a85f353282f33d212c6b44b52db/pyobjc_framework_phase-12.2-py2.py3-none-any.whl", hash = "sha256:e8b85d7b2743b61f8920ad044f7e8370de80e5a4ca4f85ff42b6047b04e5f821" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-photos"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/7e/a9/3845e4976742f15baf7decd63202b16326196596ee95271eb1eacbb388ed/pyobjc_framework_photos-12.2.tar.gz", hash = "sha256:126d2d2aa78ea0e704db807ee2c7f7b418d8a859a0cb8df6ee31e66fd371c2ab" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/b7/39/2195f0457e021692c80db6b31334d26fa08db64ced2751040384de8b5fca/pyobjc_framework_photos-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:2f3caf43105e0d142abc6b10bb851dc107158682d9dbb612d053cee80385f0f7" },
{ url = "https://mirrors.aliyun.com/pypi/packages/67/ca/3c305cec64c242f2f66533650ec955c93af89772a910c64c4eb4c6748ec2/pyobjc_framework_photos-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:59a9ac81947dc6140093ec2f4131842c8820f6ad9a51226c56e15b43ebacf011" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-photosui"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ff/62/9b4322779ac45e8971d613c23f19284d7c6ac2492b2384c8eaa4582e6dfd/pyobjc_framework_photosui-12.2.tar.gz", hash = "sha256:c85ebf27f523dca57fadf767a4d72739c760d7c7236f5e5c8fbe4f80e696c9fd" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/99/c4/c2f9cc556bde2e1e85cc20103ef34705da6e695b542330d5fff7eff17cdd/pyobjc_framework_photosui-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:2630d9f61df8f786b1a42a30872e856be2ae7a2fd21c995c5d7d88705cc784ed" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b5/68/f9696cd18c479e67b9b911c5a8d7243034fe36c6bf6482add22ae1076dc0/pyobjc_framework_photosui-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:9c3e0cdbca02ee749b3d56ff332a4a69b8d378550131df4e4003733c6331728d" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-preferencepanes"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/0b/86/73de5b65736cea66bb0159d9a37c007abceee8ecd3a059211241aa2d77cb/pyobjc_framework_preferencepanes-12.2.tar.gz", hash = "sha256:4928aa6fb30b24af3aa7379125ceca91687e97b08715909689d549a0a73266d3" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/d9/5d/b75418c408c5e969352307b5fd4d739e55b3618cf94e729ba9768e453f2a/pyobjc_framework_preferencepanes-12.2-py2.py3-none-any.whl", hash = "sha256:ccd142fb1d26f20e660651008fce1440f0abc1897bd5ffb5546cd77e89bb31a6" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-pushkit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/e1/40/23d8b7065434179984df6f4bb18c6ead5400ed2f85b7080282c600106022/pyobjc_framework_pushkit-12.2.tar.gz", hash = "sha256:8f6b6879920bf9650f2288adb0dc8f809a807825c29136b6492561712234b0b3" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/fd/73/d315aac0efb5466a0235ea69aefac9883add2f5fb90ee363372abeaab299/pyobjc_framework_pushkit-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:358e5e381d37fc1199d0e18eae0ec9bc88145545c985a75ffae37918c0634a3c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/89/6f/8ee291088ec4b6469b9b51674d0f160929395164483136cccf2569bb85c2/pyobjc_framework_pushkit-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:11a5efd1e08cce341b0882cf4b47ba120f288c04594d714c0149c2a357518b05" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-quartz"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/91/a3/5ae4c90c13999b46315f549694f25c374c48a9f7ab18f98ace6e74f4a5c1/pyobjc_framework_quartz-12.2.tar.gz", hash = "sha256:b343395d4790323b0376fe20c83ac468510ba19f65429323ca211708c939d107" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/96/56/670a847a3a8ee2799f405b876a2f20914f22b4865f1d8157169095c21d94/pyobjc_framework_quartz-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:214c19aadfd100d9202994a22fbced804f7d60f8473de6f292111cc1668f9373" },
{ url = "https://mirrors.aliyun.com/pypi/packages/35/ef/598bd4d1fb796305648c03667938f08bb59ed4e0bcdc1591fd2c6238abf2/pyobjc_framework_quartz-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:4e0634ee9782e480587a074d1d08867fa7ef0d845c2f6cbaef6a48b7d2c3899f" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-quicklookthumbnailing"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-quartz" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/75/c7/e19792f6473714f4144ff2c579f2a231eeb81e405388140471f9f6ffaa4c/pyobjc_framework_quicklookthumbnailing-12.2.tar.gz", hash = "sha256:b0104142f0eb950e849ea25f40c01715069e1bdb3de69949c1b4732c8f5c693f" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/48/5d/012d04a9fe4117f727e0436b7b6c9a30f780e04ee51f1d6e28bb6d1a8700/pyobjc_framework_quicklookthumbnailing-12.2-py2.py3-none-any.whl", hash = "sha256:a60bdc283a44a70ad7a12840010b76d42ffd580c4bffb630da3874b229d30a74" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-replaykit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ad/28/9562e12a0f13ba9c03ec59694af026500265cba2bc71468bf5985230ff3e/pyobjc_framework_replaykit-12.2.tar.gz", hash = "sha256:f971842973fad039642a67e3432b864e8bed815228d67d07b52174fceadca86f" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/f8/e3/bf9e3cc9fb4f3629b42e510f40539e323af43597601bf1ae5275a8468572/pyobjc_framework_replaykit-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:ae67c235356b4442dd0765ba3b911acf6f597894ad27b6b6c322b55ea501be6e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/03/ad/a34116524dd3d7cae22cef8125a897c4533b9dff9236cf4e00f06ca01012/pyobjc_framework_replaykit-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:2040fe7d3da7ac485c32f620bc0a2ffb8c5c1052ca67e2d36db78533bbd84e9c" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-safariservices"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/9f/e5/8caad7cc12d7452f9f78f931bec253f04577c6a876a9884834572223462c/pyobjc_framework_safariservices-12.2.tar.gz", hash = "sha256:35f76589201c9857f36fcae027c79ae2719d4463e1bb35996f78de9f570a0a58" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/83/56/c3bf80c5abf766078fc6815a91c287a72772eb675eb538fa8e3aad426dd5/pyobjc_framework_safariservices-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:d7883793626c82eb645c99b719cffa856f8fa6d2bc19ce9fabedf5611888b7e9" },
{ url = "https://mirrors.aliyun.com/pypi/packages/5c/19/bbd38d764573a907e9287aec7fd065904b136042a6f4d8d9b488cdb67640/pyobjc_framework_safariservices-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:f97571814b69ca81baa3570a4338e9c7679a7505c5f05b28938890ed71d95ff6" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-safetykit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-quartz" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/28/c2/7a6c1c356ba3d9164357c353142e6d9354f999b45f204c57c5f2162a47dd/pyobjc_framework_safetykit-12.2.tar.gz", hash = "sha256:3161c4cd35261615ea38e64bd1db9de19c66d18d4d350bbe7d4eb4f69207e256" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/a0/c9/f33171cb2129b39fc8ce4a2dfd5ccb9d5fd7fcff795b90efa85462490609/pyobjc_framework_safetykit-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:93020a92a64dba726e8e1fb03079b4697d49e67f0338b33a63eab76a1a797033" },
{ url = "https://mirrors.aliyun.com/pypi/packages/54/1d/2abc135c5b8674511bf72a194e2ee7d9fd23cc37ad6a9a8884451aa2a6b1/pyobjc_framework_safetykit-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:6dee8522c45cca1017ebc52198a89d81cf12b4e1fc751e8fab87059326b9c9d2" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-scenekit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-quartz" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f8/c8/963293a476d7c80f60d93e3da2b1f6016913e827fdde63d3b9b7ec96164b/pyobjc_framework_scenekit-12.2.tar.gz", hash = "sha256:de58706317e567d0a5f257f9a4a66fafa037c5c8ef79051c19066b42477e301f" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/a5/e1/c117e5db66ceb6c13f0bb44c40210749e2d9c61eb5d77d50e731dceb8865/pyobjc_framework_scenekit-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:0b1795cf860b19fb04116236b6de3a46ffac9f8b03c46863128d36207cc53a40" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c4/44/645de9525dbd6258954b8001336df1f3ae85559acae9a119049089975270/pyobjc_framework_scenekit-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:0d0bcea0b7a4a13046a666a5b316c1b3ebfdd90a63226fcd5b7943fd9ee9d615" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-screencapturekit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-coremedia" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/d2/e5/469c73f6ba9bf9ae6d4a93aa23838f288dabee88555402754dee000ec884/pyobjc_framework_screencapturekit-12.2.tar.gz", hash = "sha256:7f45f2e170b97dfde3e6c7d3b867ec3ec03caf3aac9259fde4ba0272d912b00e" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/73/5e/e64b2a94e8f950dc2418a274049c01ad3f8f3f54fbabe6912c72a72dff5b/pyobjc_framework_screencapturekit-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:ca8378bf217d3ac48db0bb62218b8bcf6b69bd35e1ba66608bec8c31e0f0fbf5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/21/25/892fe4f54b8ccecbcce6f05c98e377021ec4ada9b226b18523ac97c04025/pyobjc_framework_screencapturekit-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:c1be8a80eaa2a929cde641526f81e7452ff156234372d27c5084df472a74843a" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-screensaver"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/45/47/475b9d9462fb72f264c600178905e0f149d0dc3ac549940477cc582aa24c/pyobjc_framework_screensaver-12.2.tar.gz", hash = "sha256:2aefc58e53b42c33d0d06b4c846147c7b15b065254c588bef0ce8871ccf6d9d7" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/aa/cc/168c45769e522b7982702c2b2af9891036e7653d8bc50a3871f529c9fb8a/pyobjc_framework_screensaver-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:e4a2e2205effaa1dcbf5365e25011b057ae70deadeb29b0782c0af6dc2592f7c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3b/bd/f0ee37d17d48f64c1232cf89fcac33ba56454be63062ca50150ff0002799/pyobjc_framework_screensaver-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:921dc983e81b0aab7f102bcd115b4c5cbc347d251b00167ca58ccec7c6ba10cf" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-screentime"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/7b/e1/acd5b9635c905305cdd367d1e5e06f71db13c44ac3384f0241845b180272/pyobjc_framework_screentime-12.2.tar.gz", hash = "sha256:0f708bf1707f2db5c4d13397471bd13cb8c1516ea204d7c4cca8f448295caa4e" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/74/a6/2bdf31496600ffcbbb958910e03726e238f6a0ac6a7f3862cdbc28cab9c1/pyobjc_framework_screentime-12.2-py2.py3-none-any.whl", hash = "sha256:ef463bf9cf66ab2679a6fdc3da0a51bca3050bee4ff7614ac714edaa13e36003" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-scriptingbridge"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/2b/81/59528ce5c27667c0effcb6af24d039767df0faafe3ad006f154a82d270ef/pyobjc_framework_scriptingbridge-12.2.tar.gz", hash = "sha256:b1420622d923cfe6218904a6a997eae4984653d32554b16374690c347f8ab55d" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/9b/5b/ed96c67b88ce1408f1778d1235a7c3c354980e4c925ce53bfbea0b424c78/pyobjc_framework_scriptingbridge-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:ea7944907fe696a8192f2a9861e148b9a0161ac3415ab6ce6e76230181a10264" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c0/57/9658c1a8576038e571308c3f21a6f38050808a7e379b1149c9b55abed746/pyobjc_framework_scriptingbridge-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:8dfab9213a60840f15d0ebb2572237fea569ff69d2a593f524e8c7074b6cf657" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-searchkit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-coreservices" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/87/f1/2240da12e22d6e7273f60bc7fa7ec160ab67d8a3c7801010d480f661a16d/pyobjc_framework_searchkit-12.2.tar.gz", hash = "sha256:aec05c1fa302e11a8e1e737a67ed8ce2bf1e04e9e249796b3b923e90e7b07b9f" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/47/ef/4e69b33a88e31e6d4e2ef9d266771b81bd20fce0eeffc1fa7c79ce9b3ec7/pyobjc_framework_searchkit-12.2-py2.py3-none-any.whl", hash = "sha256:eb004bbc4522e7e7a8ee3751539056e18c204fd591ac8a2ffa6a8146117ab3fe" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-security"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/e3/3c/7778dd8e196373feabc54841b87e495148da3fe20d305f39397546afaaee/pyobjc_framework_security-12.2.tar.gz", hash = "sha256:ef4d2d852a09360929e284c6f355964d84ac88b170207de1ac299fa1e1c33e40" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/33/ea/20d93a62d2f3d54dccac9e3d9de617d27c229ccce42b8bd0d7f4bb4461aa/pyobjc_framework_security-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:703b8544d2623a854f438b13ef4561402e6228eacfb0f0b2cbf28d2704815f44" },
{ url = "https://mirrors.aliyun.com/pypi/packages/17/a7/9725d152c8f23c6f2d3bb7fc2c2c9bbcbc7908a91de935c80fc5ed54c6c5/pyobjc_framework_security-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:10f03ed38e915f823640cbc19d5a921ee29ae266f0d76a2910c53fcd7cd61447" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-securityfoundation"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-security" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f6/20/7fc0158fb92a894699ecf510c3be138c249817276c8346e70fd2a8868371/pyobjc_framework_securityfoundation-12.2.tar.gz", hash = "sha256:03d51d2945f4ceeb7f5fe60e28f9f18f9e96797152467d523d374f6f242eb89f" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/bb/86/f3fe3bea55f684182e65a4762204802c5fdd7f0b67db77f5dff2a2b3b2f6/pyobjc_framework_securityfoundation-12.2-py2.py3-none-any.whl", hash = "sha256:7e1c8307799cc819cf5891ed046b78eae4907e267592456688fe15f5892ec16f" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-securityinterface"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-security" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f4/21/d164e409e47ae4460d9320d21de17c825017731a870dee1a2905fd187127/pyobjc_framework_securityinterface-12.2.tar.gz", hash = "sha256:096ea141b84f5128d367f4a7800073c801200bde86a451c735708137a4f23183" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/e0/6a/39f94ce520d5661866ba15c69264eec1af67ed8e757d5ba9c25ddf0c5577/pyobjc_framework_securityinterface-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:8741dff8c98a9cf585e5dd582c3fc2df124a0b5b885e9c128a53afb960b6e4fb" },
{ url = "https://mirrors.aliyun.com/pypi/packages/5e/9d/a18302f5aac77c33b8f3e6e6b529d3cca2d934c893c880b756c8bd6434d5/pyobjc_framework_securityinterface-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:d1e56bce8179bb4b5e71986162491423173c6178663e5ca9f1f5c4ac0a493adb" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-securityui"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-security" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/c6/f0/bd802a0bca727076fc117b57046a953bbf57cb41656fda917ebf066f287f/pyobjc_framework_securityui-12.2.tar.gz", hash = "sha256:9ce580490d7be95d7ef4b6cfdb9af80b62be486932a89aa402accc86596df934" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/92/a5/db3bc24bd588cb704bb7db16f0f73177b0488c7577d676ced3932f1f41ee/pyobjc_framework_securityui-12.2-py2.py3-none-any.whl", hash = "sha256:d6cc86e9c039e8a1ac0227947277fb2b46ca2b9f9bb1fb7f7b081a3717ac045e" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-sensitivecontentanalysis"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-quartz" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/03/f9/e18f459b2957f208718e7ff6a9f2715ad66509bb3bb1d4988d4c5820941b/pyobjc_framework_sensitivecontentanalysis-12.2.tar.gz", hash = "sha256:796d6cd3696daa7ac9116b35e7b2d520e86b687a8433443b755db8fd0cdd8315" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/e1/37/3c92718706b58f76eeacbafc5d6ba1b09a5ff684cd8f47ee7b53dc139d70/pyobjc_framework_sensitivecontentanalysis-12.2-py2.py3-none-any.whl", hash = "sha256:5beb7d718e6dea0b17c14560ebaa3474f7aa349e412f974573cfae84cd7121b7" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-servicemanagement"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/b5/d6/c31e62d53305b30110cae00013b0947b5cec17ee77afe3e1376d5fe1ea1c/pyobjc_framework_servicemanagement-12.2.tar.gz", hash = "sha256:9c7b698f4354a36ad0448dccba57e724ac682a1276eef3971b8d85d7ac7a6488" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/09/67/d02639a189926546ddd2afd5015c0a5574e8310c2a59633bff0ad2b90b2a/pyobjc_framework_servicemanagement-12.2-py2.py3-none-any.whl", hash = "sha256:fa8b9d3bcfd0d2e6feb0884f3a95fda9cd681baf0b0ca457ec3f611de9439f7f" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-sharedwithyou"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-sharedwithyoucore" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f6/bb/afb864c9b6835bae72d55208e52d79734a88d597edc89c437a717da34ef5/pyobjc_framework_sharedwithyou-12.2.tar.gz", hash = "sha256:215aefe1baee8d31dce55b38bd653cdecf0a74dddcf89277d64fcb8d6beafb9d" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/22/2d/5df7613183733cb6d8bcb811d3b423cf97a24a12803cda0caaaa5018e5c1/pyobjc_framework_sharedwithyou-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:1e6d829807fb4edd2eaf5752319649b99a4574d359cd0742c71c7ec0037fec82" },
{ url = "https://mirrors.aliyun.com/pypi/packages/16/cc/5b5bb308f516f9ad75089595f3bf06eb08e1641cf4b66747473cb45bdf12/pyobjc_framework_sharedwithyou-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:ee9b7aba9c09b8f7c75982cfc714f16d2c8db9dff54eaba689b4ee51f84c2124" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-sharedwithyoucore"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/b0/02/4b1809c63319c277d88542576f86b7d7abcffd57ccd876d47d4623d6d5fb/pyobjc_framework_sharedwithyoucore-12.2.tar.gz", hash = "sha256:212bd551676cf0ae791eb8d7b5f6a4b10090e2d407721124d691e3020f709e51" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/34/3e/913bc40fce99ee370c940261eb59d65bd4f9961354c5059bd4d5612e32b2/pyobjc_framework_sharedwithyoucore-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:3cc10c8a03892a3da8bff22235e31baa24a931f5bcd1994fc0124a197f14f70e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2e/94/67b218f72f4a6be6926691fc85266437edd7fcc837cec755a43bc0daa206/pyobjc_framework_sharedwithyoucore-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:0286e5dfb8bc9530de31b0e686d052019d91e17fff3ea35c0338ac62e23659b6" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-shazamkit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a1/c5/5cc4e2c0cb5e3fb3a3e3a4eb75aa85da70c5491125efef25e581516ec108/pyobjc_framework_shazamkit-12.2.tar.gz", hash = "sha256:9bfd2790b331b36ebdd9f80dab1fcc80b816ee19a1fb5ae981e5b4a8fb2e7084" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/f0/e4/837c888b09e330599601446b0da9ec3101ee1ad4250113c5f4a040ea76d1/pyobjc_framework_shazamkit-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:05b0f1efc47471a2962f947dcf764355086a37061307880ca4853df3d7d202df" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a2/48/ca405138c3deae6a18b8750ab8b1e688add932e149ce234bb132b999e034/pyobjc_framework_shazamkit-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:ef389bb99ac4c74741fa1b08f4b08721cbeed3522f2d8e167a9105545d5d29f6" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-social"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/9e/8e/30527d2219e08e0d042c1190ffb64425aca051701f50823d5376a22bf573/pyobjc_framework_social-12.2.tar.gz", hash = "sha256:0f5e8c3e6bfd36f8c552b58b42d8d1ff4c8bf18c74776cdc3e830c272a904118" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/24/93/e8f2405693ca24878b1fe1cf58d4aa58b293f4e774af64a56aeba904b33b/pyobjc_framework_social-12.2-py2.py3-none-any.whl", hash = "sha256:d012f52721d694000b37999b2fed213332bf89e8a682f4d29656db4d8c6c7087" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-soundanalysis"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/97/0a/19357da779f1b8942f69acfaa3f4568c0b117d6c1b5820af2a50d25dde31/pyobjc_framework_soundanalysis-12.2.tar.gz", hash = "sha256:37dee57e3f75121b690550601fc136e0931f3be56a87357cdfae7696886850fc" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/23/8b/fb4db5583f5be6a5537baa1ecc39ae56f2bc1bdd6fc8dc7912e7c48b36fa/pyobjc_framework_soundanalysis-12.2-py2.py3-none-any.whl", hash = "sha256:d669deee79636bc6858e40039d5c9b4cca47af7b4cf042c787c52f39b5b5ab40" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-speech"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/5a/d7/b4c297150b863946fc4064dd927445d0539140524629e4d0b12f704577c0/pyobjc_framework_speech-12.2.tar.gz", hash = "sha256:14058f3ab48e29a68de5925297754a8bc6157b60c17cb24b3e8560290769e9aa" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ce/bb/4262b2565e6fd0137b07b661a828acc48d65c6280124755feeac8fd1964b/pyobjc_framework_speech-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:4647b2775c3bdd3016fb5cce155dd03ca7a8f04902211868a68099511e8ea9f7" },
{ url = "https://mirrors.aliyun.com/pypi/packages/96/48/4071527182ef72ee5e31158cfbdf11d6f97265708e94a0a1ed138f2620e0/pyobjc_framework_speech-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:df58c5d6ec1ac5425565a1544659c8c45c78b0f7984ac50ff4590b5b6b872a5d" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-spritekit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-quartz" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/de/1d/84896e2233d2f90444e3be44282ad6e8600b485ce96c37566ed5bd255204/pyobjc_framework_spritekit-12.2.tar.gz", hash = "sha256:75e8fd8040f77c7585247004b9c59971de84cbcb707070f391c8508888750565" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/9e/39/abc521893bd60cf5b88ece0091e83fd30717c58a2dbce712d9d9da2e5b3b/pyobjc_framework_spritekit-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:d8ce3c810728a0a54f0af3063ed7ae45a8d0ae6c5a5ba405d02d08cf2c2bc6e5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ec/44/aaa0c92b8bf0b9b32c13fc62fdf506e3a868f3451670291441fe008bf9e9/pyobjc_framework_spritekit-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:a8a68b20f5377ee1530f752bb693388a3dcc085b1c2cdaf6930427d1cf5791a4" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-storekit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/29/ce/3c5e9ee38040e4770af6aafa0fd11080b34a20dc2c718d2524db5b048bce/pyobjc_framework_storekit-12.2.tar.gz", hash = "sha256:f50699484da541b9d028503441bcb745eeff0becadc75556783aaf76b8764975" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/1b/b7/16adfa4c5aedad949ea10a751c60e343745afe889706bb8d71491ff57732/pyobjc_framework_storekit-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:94d2daae9e5166cfa9d694acb2e51bf1d3856cf7ce6c74a84e13d80dbe6915ec" },
{ url = "https://mirrors.aliyun.com/pypi/packages/39/c9/bcf7f4c1087b7f83d190335d2149e8ecd5591064419ea820dac09b47bf4b/pyobjc_framework_storekit-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:922da8a72885d79c3c30dd3d36e9a966dcdd642c08f258b4b099a3f0e3ded643" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-symbols"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/50/06/042965972619d440c69037105764accd700053c658dbcbc2914ba4682ab5/pyobjc_framework_symbols-12.2.tar.gz", hash = "sha256:06c00d97047ee79fad95f95380d308c6dc1200d688cf4adf954382d65f138b10" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/6d/de/168b3925a76417978e5bc5921c8ac0a278e5a672302a20ca10f18d317695/pyobjc_framework_symbols-12.2-py2.py3-none-any.whl", hash = "sha256:abc83c18ef8733897667d0b4d79400e3c38828347985bcad4aaf40b0cd61c94a" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-syncservices"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-coredata" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f3/e4/ac36cf4a79c52c865d92b148b024ee0fcda7e9efa45017cc64654558654b/pyobjc_framework_syncservices-12.2.tar.gz", hash = "sha256:354f7aab848d4f2ce055886a083d07be752ebedf5fe1dcb3cbedd489fd4bfa05" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/60/1c/1ef02418ff43f53ac9ef49035b34fcdf664fcd5632d3d796cafddd01fd57/pyobjc_framework_syncservices-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:a9ff8eb0475033f24434b1dfd7f77a8c458f9ef7ed692e59b9e5a26c22692e0d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/21/ea/df2f2e636b8616ccb4c8d2b15ddd29275245ba60a7666dc9287b4fe977a5/pyobjc_framework_syncservices-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:10c1dff31d1499b618b3741956219577f2254b04df2fbf10a65a49ed3ff3dff2" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-systemconfiguration"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a3/de/30ce4e5e41d2bc965b5598ba4f3595ac3eaa088e26cc99d519f68d62c7b9/pyobjc_framework_systemconfiguration-12.2.tar.gz", hash = "sha256:017bb958c436b6aa90e04a99fe3fdce3787e8c1db80d165ff0071273bdbf242f" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ce/e3/fb4e684f873347039ea5136c0a5f7a0ae18ce7b4caaee8508e013fd47f03/pyobjc_framework_systemconfiguration-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:d0b290b3cfd77fde1f4234590afa0be7785c02ff2cd5b591ef8d07e2cd0caecd" },
{ url = "https://mirrors.aliyun.com/pypi/packages/26/0e/948a070637f7f0ffcf2dd3b5a323c77694a1ed0ae04e54dd4fddea9cfac2/pyobjc_framework_systemconfiguration-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:d4b81bb37f365891a14bb4a09d735784a0bbe2c594b89a1cb87e17f1db7abac8" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-systemextensions"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/9c/4a/49dd15416665943fe08999c3d5d8675c6dc153c720b9cc17e83215350a46/pyobjc_framework_systemextensions-12.2.tar.gz", hash = "sha256:2ec1edcd9d12451ef74969eae546b4661514381d7d60221240560ca9a743f506" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/dd/76/f9490cb2be6a6c06ba2132a50d33d368dd90a917284a712614eec818c08b/pyobjc_framework_systemextensions-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:dc3c41d413a867483511edbd3332d3a13fa0011119fadf0e4c642ea1b97b7bec" },
{ url = "https://mirrors.aliyun.com/pypi/packages/db/4c/a9e07ceece91b207e2554166675f299f05b0fe9af93868b91eaeedfeb2d0/pyobjc_framework_systemextensions-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:8c131f836a61df9064020382b6b1b80965799ae44cdb197ef4696167e0796bf2" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-threadnetwork"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ce/e6/8b1744871062c39cce64d43ab20b11344ef5f1b93993c1b737da2faa3acd/pyobjc_framework_threadnetwork-12.2.tar.gz", hash = "sha256:2d7df5938094e3aa93210a079ebf3e801a20c84a3b759573ddf05ec2c2cbf083" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/a4/7e/dfa9e914c1f192d6afdb504efc01f5aac777a73d43cfe103896bb5072ee2/pyobjc_framework_threadnetwork-12.2-py2.py3-none-any.whl", hash = "sha256:08c7a03e11c60dca7b3f219db744986b90b8797d61b6842e0c0bad192e723f48" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-uniformtypeidentifiers"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/64/54/e270da2b0b4ab1fb0dc7f4e616d1a6e141583f88409c0742f6ebfcf7a064/pyobjc_framework_uniformtypeidentifiers-12.2.tar.gz", hash = "sha256:19cc82f1bbf3bc0999597779711422f6c9d7340634d699ae64d9bc7c0d79f984" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/da/19/4f314697bc8f519fe6505afe51d83d7fdcae93239b7bab5941f4bebc1f9e/pyobjc_framework_uniformtypeidentifiers-12.2-py2.py3-none-any.whl", hash = "sha256:f140a378cfe6a8ca47ce3b04fd5a4c4bec1fcbedac8acc87e2c18985bb805203" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-usernotifications"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/5a/c3/f1d43ac73b5fce42a00bcb68ac53edd7180217df60a25985e2477c9c389b/pyobjc_framework_usernotifications-12.2.tar.gz", hash = "sha256:b7773e84fe40e746f706055267ba2b1dc9e0a34e6575719f1e515354f72583a0" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/dd/e3/a962cbd3739cbb01b2b5697e336babb15307fb473500524bd7e730446b7e/pyobjc_framework_usernotifications-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:dbc51fc214446f622b4670bca7efb8969ad84cdee21672da3c872fa3aff5aff5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/dd/8e/2a44292fc4e96682054f74a1aab1c303fa5caf5820e5cd67b0d23b52750f/pyobjc_framework_usernotifications-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:e329301b42181a95c2ac5f71491529ac1260afa9b26b16a104db275bbafdcb18" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-usernotificationsui"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-usernotifications" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/4b/5f/d58600259b0b99ff3745eef91570d0ced05fa1b26083f2846082acc15249/pyobjc_framework_usernotificationsui-12.2.tar.gz", hash = "sha256:8c3298cddd27be794ad46a2783304b6fb1f0ad4d65933c34ba3e43e58ecdb406" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/c6/29/f4413e1941e153a770b2d85faee4d3a4f9be67b69f5823f63db8dd091356/pyobjc_framework_usernotificationsui-12.2-py2.py3-none-any.whl", hash = "sha256:a5d92117c2d18e2b6365f5e39daf45a68491a66ac9103b1bc94df43b8868c256" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-videosubscriberaccount"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/42/3d/cace0ae1dd453f9da495311675d4fabe7adf89e94e7b99e26f45ba0fa713/pyobjc_framework_videosubscriberaccount-12.2.tar.gz", hash = "sha256:821f5485545238fd997dc26aff8ab2c838b43d664a7ec6a030ba67b09c2b799d" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/d3/03/f3b0c4ed85b4883d5fcd48ba2efd807507c3a5ce7c3aa7d4366dce81f503/pyobjc_framework_videosubscriberaccount-12.2-py2.py3-none-any.whl", hash = "sha256:b158b03e1432dc229fabc96451f5ac8db5ab228792354f0f26419a973777c700" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-videotoolbox"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-coremedia" },
{ name = "pyobjc-framework-quartz" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f3/2c/af5a91d12bb265697fb13ccdbac7068e04e8f40c36139b314ea477982aa3/pyobjc_framework_videotoolbox-12.2.tar.gz", hash = "sha256:33d0838f706c771667b0be5737036d303b677bb243ba64df9ea2fa6860ea9f76" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/c9/08/48b4af0452930b1865dee4e847113d5b6d2530a19ad111b6f551d8b5bd8a/pyobjc_framework_videotoolbox-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:3393990f9f566dadada78b06c134bcefcabd50731b5322110759e0d26d9ca280" },
{ url = "https://mirrors.aliyun.com/pypi/packages/48/11/d111f5d93f88f0d4d7150d318a6e089b0e08196cc29bdcae3c6af7c9891b/pyobjc_framework_videotoolbox-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:06dea99b364c3628037332c5fbffcb341f32a4cc7acd0ea4813a83b78fe1f70c" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-virtualization"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/1a/b4/16c8280c717f393e524e32de93bcb18952ca852ebbf5aaed38dac68d2e01/pyobjc_framework_virtualization-12.2.tar.gz", hash = "sha256:021249f0cd72e4756dfba096bda0e9ebd9294ba7738f1bdbb2742e461ac5671e" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/46/90/a2e2a3676286359dcf4d5898652e904925006dba0eca959759472be37651/pyobjc_framework_virtualization-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:48667f1318bdc3e3599123dafb1d9df3369a6d8d3a64fb36f9eaf86ef2ea4808" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9d/22/3e04b58c4a1ecc27518a8c5b8ca13cea7b74106928b1bd130a0564cb54a6/pyobjc_framework_virtualization-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:68e8989bf2cfb294116cdfd583d6ac092f9209ccca2af5b79fb697adbdae4653" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-vision"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
{ name = "pyobjc-framework-coreml" },
{ name = "pyobjc-framework-quartz" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/49/51/aed3761cfa2f0335a8fc4cdfa04c54b3fece5242745d5dd2370f38efcff5/pyobjc_framework_vision-12.2.tar.gz", hash = "sha256:0e90244f98ede5ec16ac129ed4bbd7cf0ec07247bac6f06e9cb612db23a68a3b" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/79/34/9c92c1d70a98818edb4e113f771bb264c2e7f08ec6bba501e0196c500dc4/pyobjc_framework_vision-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:df4d7d3ccdd723334ab62e6e2051d04e3554724be90102db353874938f101e05" },
{ url = "https://mirrors.aliyun.com/pypi/packages/16/a5/1cf2213da57b1941f788baf0505c3cc4c98fbbdcdf1ac3a19290331ca4d1/pyobjc_framework_vision-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:76737fa3440c504993995537643b6529d47c1affecba426f00c716cd93a07a9b" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobjc-framework-webkit"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pyobjc-core" },
{ name = "pyobjc-framework-cocoa" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/06/ba/0e17f151d80854285a84b3c96f48d55b0c61ae683060a52cd50084ef4c64/pyobjc_framework_webkit-12.2.tar.gz", hash = "sha256:498353fe812006f7fe2829a679018fcff3ed2684aa31c08d8a8af929ee9283b6" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/f7/32/bea045cf4e2c697ea9b68ed7d60b0e765a8b0e46d30a4eb26bf178be817c/pyobjc_framework_webkit-12.2-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:9fafcbf5a071cf9ad3fbc28155a99287ff9dc9efcaa41fe3ed3d79b6e5576450" },
{ url = "https://mirrors.aliyun.com/pypi/packages/bc/3e/d29fc44f6d346a5a97883ae76bfa030a5b6ff3b019bafd58ad8ea14588d2/pyobjc_framework_webkit-12.2-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:751f3138441c81fdddc555f0bd31f5f3ad3ec72ae2ed626af24925ce32054353" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyobvector"
version = "0.2.22"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "aiomysql" },
{ name = "numpy" },
{ name = "pydantic" },
{ name = "pymysql" },
{ name = "sqlalchemy" },
{ name = "sqlglot" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/30/b9/443d65757cdfb47d31ef4b9ed0609628ae468e52e57033051e1fad256c59/pyobvector-0.2.22.tar.gz", hash = "sha256:0bd4af46cfdfbc67e691d5b49f3b0662f702a7a42a7f7a240f1021af378e793c" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/e0/88/1583888a4ce85202d93fa03f2817681637465668e8b260ef1b9d5a39c3ca/pyobvector-0.2.22-py3-none-any.whl", hash = "sha256:4a0f5c094af7ca8242fdf9e5111e75544de0a9615491e9ec2f9d218dc909b509" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyodbc"
version = "5.3.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/8f/85/44b10070a769a56bd910009bb185c0c0a82daff8d567cd1a116d7d730c7d/pyodbc-5.3.0.tar.gz", hash = "sha256:2fe0e063d8fb66efd0ac6dc39236c4de1a45f17c33eaded0d553d21c199f4d05" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/0c/1d/9e74cbcc1d4878553eadfd59138364b38656369eb58f7e5b42fb344c0ce7/pyodbc-5.3.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:7e9ab0b91de28a5ab838ac4db0253d7cc8ce2452efe4ad92ee6a57b922bf0c24" },
{ url = "https://mirrors.aliyun.com/pypi/packages/37/c7/27d83f91b3144d3e275b5b387f0564b161ddbc4ce1b72bb3b3653e7f4f7a/pyodbc-5.3.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:6132554ffbd7910524d643f13ce17f4a72f3a6824b0adef4e9a7f66efac96350" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1b/33/2bb24e7fc95e98a7b11ea5ad1f256412de35d2e9cc339be198258c1d9a76/pyodbc-5.3.0-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:1629af4706e9228d79dabb4863c11cceb22a6dab90700db0ef449074f0150c0d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fa/24/88cde8b6dc07a93a92b6c15520a947db24f55db7bd8b09e85956642b7cf3/pyodbc-5.3.0-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:5ceaed87ba2ea848c11223f66f629ef121f6ebe621f605cde9cfdee4fd9f4b68" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c2/99/53c08562bc171a618fa1699297164f8885e66cde38c3b30f454730d0c488/pyodbc-5.3.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:3cc472c8ae2feea5b4512e23b56e2b093d64f7cbc4b970af51da488429ff7818" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d8/10/68a0b5549876d4b53ba4c46eed2a7aca32d589624ed60beef5bd7382619e/pyodbc-5.3.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:c79df54bbc25bce9f2d87094e7b39089c28428df5443d1902b0cc5f43fd2da6f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/41/0f/9dfe4987283ffcb981c49a002f0339d669215eb4a3fe4ee4e14537c52852/pyodbc-5.3.0-cp313-cp313-win32.whl", hash = "sha256:c2eb0b08e24fe5c40c7ebe9240c5d3bd2f18cd5617229acee4b0a0484dc226f2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/56/03/15dcefe549d3888b649652af7cca36eda97c12b6196d92937ca6d11306e9/pyodbc-5.3.0-cp313-cp313-win_amd64.whl", hash = "sha256:01166162149adf2b8a6dc21a212718f205cabbbdff4047dc0c415af3fd85867e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c4/c1/c8b128ae59a14ecc8510e9b499208e342795aecc3af4c3874805c720b8db/pyodbc-5.3.0-cp313-cp313-win_arm64.whl", hash = "sha256:363311bd40320b4a61454bebf7c38b243cd67c762ed0f8a5219de3ec90c96353" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyopenssl"
fix: upgrade crawl4ai to 0.8.0 (CVE-2026-26217) (#15415) ## Summary Upgrade crawl4ai from 0.7.6 to 0.8.0 to fix CVE-2026-26217. ## Vulnerability | Field | Value | |-------|-------| | **ID** | CVE-2026-26217 | | **Severity** | CRITICAL | | **Scanner** | trivy | | **Rule** | `CVE-2026-26217` | | **File** | `uv.lock` | | **Assessment** | Likely exploitable | **Description**: Crawl4AI Has Local File Inclusion in Docker API via file:// URLs ## Evidence **Scanner confirmation**: trivy rule `CVE-2026-26217` flagged this pattern. **Production code**: This file is in the production codebase, not test-only code. ## Threat Model Context This is a web service - vulnerabilities in request handlers are directly exploitable by remote attackers. ## Changes - `pyproject.toml` - `uv.lock` ## Verification - [x] Build passes - [x] Scanner re-scan confirms fix - [x] LLM code review passed --- *This change addresses a pattern flagged by static analysis. The code path handles user-influenced input and the fix reduces the attack surface against both manual and automated exploitation.* --- *Automated security fix by [OrbisAI Security](https://orbisappsec.com)*
2026-05-29 19:08:41 +05:30
version = "26.2.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "cryptography" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/1a/51/27a5ad5f939d08f690a326ef9582cda7140555180db71695f6fb747d6a36/pyopenssl-26.2.0.tar.gz", hash = "sha256:8c6fcecd1183a7fc897548dfe388b0cdb7f37e018200d8409cf33959dbe35387" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/73/b8/a0e2790ae249d6f38c9f66de7a211621a7ab2650217bcd04e1262f578a56/pyopenssl-26.2.0-py3-none-any.whl", hash = "sha256:4f9d971bc5298b8bc1fab282803da04bf000c755d4ad9d99b52de2569ca19a70" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyotp"
version = "2.9.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f3/b2/1d5994ba2acde054a443bd5e2d384175449c7d2b6d1a0614dbca3a63abfc/pyotp-2.9.0.tar.gz", hash = "sha256:346b6642e0dbdde3b4ff5a930b664ca82abfa116356ed48cc42c7d6590d36f63" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/c3/c0/c33c8792c3e50193ef55adb95c1c3c2786fe281123291c2dbf0eaab95a6f/pyotp-2.9.0-py3-none-any.whl", hash = "sha256:81c2e5865b8ac55e825b0358e496e1d9387c811e85bb40e71a3b29b288963612" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pypandoc"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.16.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/0b/18/9f5f70567b97758625335209b98d5cb857e19aa1a9306e9749567a240634/pypandoc-1.16.2.tar.gz", hash = "sha256:7a72a9fbf4a5dc700465e384c3bb333d22220efc4e972cb98cf6fc723cdca86b" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/bb/e9/b145683854189bba84437ea569bfa786f408c8dc5bc16d8eb0753f5583bf/pypandoc-1.16.2-py3-none-any.whl", hash = "sha256:c200c1139c8e3247baf38d1e9279e85d9f162499d1999c6aa8418596558fe79b" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyparsing"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "3.3.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/33/c1/1d9de9aeaa1b89b0186e5fe23294ff6517fce1bc69149185577cd31016b2/pyparsing-3.3.1.tar.gz", hash = "sha256:47fad0f17ac1e2cad3de3b458570fbc9b03560aa029ed5e16ee5554da9a2251c" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/8b/40/2614036cdd416452f5bf98ec037f38a1afb17f327cb8e6b652d4729e0af8/pyparsing-3.3.1-py3-none-any.whl", hash = "sha256:023b5e7e5520ad96642e2c6db4cb683d3970bd640cdf7115049a6e9c3682df82" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pypdf"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "6.13.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/15/d9/9d12fa0d9660d03320725ff686c961b645a4218940a82296e1272d9e1ff0/pypdf-6.13.1.tar.gz", hash = "sha256:4841d8a4c1589e5833915dc0c7ddfacff80a2e0bcbeb5d1e681fecaa1674b03a" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/fe/dd/8f03e0a5788a5d1feb4550617c3e6db5e9099eaee248a3e482ddaeacbbb0/pypdf-6.13.1-py3-none-any.whl", hash = "sha256:e555e4ce3f561ef069307622f1374136ba964ca6ca24f24158701decaf83ed9b" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pypdfium2"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "5.3.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/18/83/173dab58beb6c7e772b838199014c173a2436018dd7cfde9bbf4a3be15da/pypdfium2-5.3.0.tar.gz", hash = "sha256:2873ffc95fcb01f329257ebc64a5fdce44b36447b6b171fe62f7db5dc3269885" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/e3/a4/6bb5b5918c7fc236ec426be8a0205a984fe0a26ae23d5e4dd497398a6571/pypdfium2-5.3.0-py3-none-android_23_arm64_v8a.whl", hash = "sha256:885df6c78d41600cb086dc0c76b912d165b5bd6931ca08138329ea5a991b3540" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3e/64/24b41b906006bf07099b095f0420ee1f01a3a83a899f3e3731e4da99c06a/pypdfium2-5.3.0-py3-none-android_23_armeabi_v7a.whl", hash = "sha256:6e53dee6b333ee77582499eff800300fb5aa0c7eb8f52f95ccb5ca35ebc86d48" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c2/c0/3ec73f4ded83ba6c02acf6e9d228501759d5d74fe57f1b93849ab92dcc20/pypdfium2-5.3.0-py3-none-macosx_11_0_arm64.whl", hash = "sha256:ce4466bdd62119fe25a5f74d107acc9db8652062bf217057630c6ff0bb419523" },
{ url = "https://mirrors.aliyun.com/pypi/packages/62/ca/e553b3b8b5c2cdc3d955cc313493ac27bbe63fc22624769d56ded585dd5e/pypdfium2-5.3.0-py3-none-macosx_11_0_x86_64.whl", hash = "sha256:cc2647fd03db42b8a56a8835e8bc7899e604e2042cd6fedeea53483185612907" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a1/56/615b776071e95c8570d579038256d0c77969ff2ff381e427be4ab8967f44/pypdfium2-5.3.0-py3-none-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:35e205f537ddb4069e4b4e22af7ffe84fcf2d686c3fee5e5349f73268a0ef1ca" },
{ url = "https://mirrors.aliyun.com/pypi/packages/df/10/27114199b765bdb7d19a9514c07036ad2fc3a579b910e7823ba167ead6de/pypdfium2-5.3.0-py3-none-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:b5795298f44050797ac030994fc2525ea35d2d714efe70058e0ee22e5f613f27" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b4/d7/2a3afa35e6c205a4f6264c33b8d2f659707989f93c30b336aa58575f66fa/pypdfium2-5.3.0-py3-none-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:b7cd43dfceb77137e69e74c933d41506da1dddaff70f3a794fb0ad0d73e90d75" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a2/f1/6658755cf6e369bb51d0bccb81c51c300404fbe67c2f894c90000b6442dd/pypdfium2-5.3.0-py3-none-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:d5956867558fd3a793e58691cf169718864610becb765bfe74dd83f05cbf1ae3" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f5/34/f86482134fa641deb1f524c45ec7ebd6fc8d404df40c5657ddfce528593e/pypdfium2-5.3.0-py3-none-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:b3ff1071e9a782625822658dfe6e29e3a644a66960f8713bb17819f5a0ac5987" },
{ url = "https://mirrors.aliyun.com/pypi/packages/09/34/40ab99425dcf503c172885904c5dc356c052bfdbd085f9f3cc920e0b8b25/pypdfium2-5.3.0-py3-none-manylinux_2_27_s390x.manylinux_2_28_s390x.whl", hash = "sha256:f319c46ead49d289ab8c1ed2ea63c91e684f35bdc4cf4dc52191c441182ac481" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a5/67/0f7532f80825a7728a5cbff3f1104857f8f9fe49ebfd6cb25582a89ae8e1/pypdfium2-5.3.0-py3-none-manylinux_2_38_riscv64.manylinux_2_39_riscv64.whl", hash = "sha256:6dc67a186da0962294321cace6ccc0a4d212dbc5e9522c640d35725a812324b8" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ce/6c/c03d2a3d6621b77aac9604bce1c060de2af94950448787298501eac6c6a2/pypdfium2-5.3.0-py3-none-musllinux_1_2_aarch64.whl", hash = "sha256:0ad0afd3d2b5b54d86287266fd6ae3fef0e0a1a3df9d2c4984b3e3f8f70e6330" },
{ url = "https://mirrors.aliyun.com/pypi/packages/af/39/9ad1f958cbe35d4693ae87c09ebafda4bb3e4709c7ccaec86c1a829163a3/pypdfium2-5.3.0-py3-none-musllinux_1_2_armv7l.whl", hash = "sha256:1afe35230dc3951b3e79b934c0c35a2e79e2372d06503fce6cf1926d2a816f47" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2a/e2/4d32310166c2d6955d924737df8b0a3e3efc8d133344a98b10f96320157d/pypdfium2-5.3.0-py3-none-musllinux_1_2_i686.whl", hash = "sha256:00385793030cadce08469085cd21b168fd8ff981b009685fef3103bdc5fc4686" },
{ url = "https://mirrors.aliyun.com/pypi/packages/14/ea/38c337ff12a8cec4b00fd4fdb0a63a70597a344581e20b02addbd301ab56/pypdfium2-5.3.0-py3-none-musllinux_1_2_ppc64le.whl", hash = "sha256:d911e82676398949697fef80b7f412078df14d725a91c10e383b727051530285" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a1/77/9d8de90c35d2fc383be8819bcde52f5821dacbd7404a0225e4010b99d080/pypdfium2-5.3.0-py3-none-musllinux_1_2_riscv64.whl", hash = "sha256:ca1dc625ed347fac3d9002a3ed33d521d5803409bd572e7b3f823c12ab2ef58f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a5/39/9d4a6fbd78fcb6803b0ea5e4952a31d6182a0aaa2609cfcd0eb88446fdb8/pypdfium2-5.3.0-py3-none-musllinux_1_2_s390x.whl", hash = "sha256:ea4f9db2d3575f22cd41f4c7a855240ded842f135e59a961b5b1351a65ce2b6e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9d/38/cdd4ed085c264234a59ad32df1dfe432c77a7403da2381e0fcc1ba60b74e/pypdfium2-5.3.0-py3-none-musllinux_1_2_x86_64.whl", hash = "sha256:0ea24409613df350223c6afc50911c99dca0d43ddaf2616c5a1ebdffa3e1bcb5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/93/4c/d2f40145c9012482699664f615d7ae540a346c84f68a8179449e69dcc4d8/pypdfium2-5.3.0-py3-none-win32.whl", hash = "sha256:5bf695d603f9eb8fdd7c1786add5cf420d57fbc81df142ed63c029ce29614df9" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2c/dc/1388ea650020c26ef3f68856b9227e7f153dcaf445e7e4674a0b8f26891e/pypdfium2-5.3.0-py3-none-win_amd64.whl", hash = "sha256:8365af22a39d4373c265f8e90e561cd64d4ddeaf5e6a66546a8caed216ab9574" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c8/71/a433668d33999b3aeb2c2dda18aaf24948e862ea2ee148078a35daac6c1c/pypdfium2-5.3.0-py3-none-win_arm64.whl", hash = "sha256:0b2c6bf825e084d91d34456be54921da31e9199d9530b05435d69d1a80501a12" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pyreadline3"
version = "3.5.4"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/0f/49/4cea918a08f02817aabae639e3d0ac046fef9f9180518a3ad394e22da148/pyreadline3-3.5.4.tar.gz", hash = "sha256:8d57d53039a1c75adba8e50dd3d992b28143480816187ea5efbd5c78e6c885b7" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/5a/dc/491b7661614ab97483abf2056be1deee4dc2490ecbf7bff9ab5cdbac86e1/pyreadline3-3.5.4-py3-none-any.whl", hash = "sha256:eaf8e6cc3c49bcccf145fc6067ba8643d1df34d604a1ec0eccbf7a18e6d3fae6" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pysocks"
version = "1.7.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/bd/11/293dd436aea955d45fc4e8a35b6ae7270f5b8e00b53cf6c024c83b657a11/PySocks-1.7.1.tar.gz", hash = "sha256:3f8804571ebe159c380ac6de37643bb4685970655d3bba243530d6558b799aa0" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/8d/59/b4572118e098ac8e46e399a1dd0f2d85403ce8bbaad9ec79373ed6badaf9/PySocks-1.7.1-py3-none-any.whl", hash = "sha256:2725bd0a9925919b9b51739eea5f9e2bae91e83288108a9ad338b2e3a4435ee5" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pytest"
version = "9.0.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "colorama", marker = "sys_platform == 'win32'" },
{ name = "iniconfig" },
{ name = "packaging" },
{ name = "pluggy" },
{ name = "pygments" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/d1/db/7ef3487e0fb0049ddb5ce41d3a49c235bf9ad299b6a25d5780a89f19230f/pytest-9.0.2.tar.gz", hash = "sha256:75186651a92bd89611d1d9fc20f0b4345fd827c41ccd5c299a868a05d70edf11" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/3b/ab/b3226f0bd7cdcf710fbede2b3548584366da3b19b5021e74f5bde2a8fa3f/pytest-9.0.2-py3-none-any.whl", hash = "sha256:711ffd45bf766d5264d487b917733b453d917afd2b0ad65223959f59089f875b" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pytest-asyncio"
version = "1.3.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pytest" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/90/2c/8af215c0f776415f3590cac4f9086ccefd6fd463befeae41cd4d3f193e5a/pytest_asyncio-1.3.0.tar.gz", hash = "sha256:d7f52f36d231b80ee124cd216ffb19369aa168fc10095013c6b014a34d3ee9e5" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/e5/35/f8b19922b6a25bc0880171a2f1a003eaeb93657475193ab516fd87cac9da/pytest_asyncio-1.3.0-py3-none-any.whl", hash = "sha256:611e26147c7f77640e6d0a92a38ed17c3e9848063698d5c93d5aa7aa11cebff5" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pytest-base-url"
version = "2.1.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "pytest" },
{ name = "requests" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ae/1a/b64ac368de6b993135cb70ca4e5d958a5c268094a3a2a4cac6f0021b6c4f/pytest_base_url-2.1.0.tar.gz", hash = "sha256:02748589a54f9e63fcbe62301d6b0496da0d10231b753e950c63e03aee745d45" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/98/1c/b00940ab9eb8ede7897443b771987f2f4a76f06be02f1b3f01eb7567e24a/pytest_base_url-2.1.0-py3-none-any.whl", hash = "sha256:3ad15611778764d451927b2a53240c1a7a591b521ea44cebfe45849d2d2812e6" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pytest-cov"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "7.0.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "coverage" },
{ name = "pluggy" },
{ name = "pytest" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/5e/f7/c933acc76f5208b3b00089573cf6a2bc26dc80a8aece8f52bb7d6b1855ca/pytest_cov-7.0.0.tar.gz", hash = "sha256:33c97eda2e049a0c5298e91f519302a1334c26ac65c1a483d6206fd458361af1" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ee/49/1377b49de7d0c1ce41292161ea0f721913fa8722c19fb9c1e3aa0367eecb/pytest_cov-7.0.0-py3-none-any.whl", hash = "sha256:3b8e9558b16cc1479da72058bdecf8073661c7f57f7d3c5f22a1c23507f2d861" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pytest-playwright"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "0.8.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "playwright" },
{ name = "pytest" },
{ name = "pytest-base-url" },
{ name = "python-slugify" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/58/ef/172eb8e23c80491fc72f1401c72f9305663873649351306a38b18406b0c9/pytest_playwright-0.8.0.tar.gz", hash = "sha256:7888d4a2443160c82e0c506c437076679f86b36d1910427250d90fbf1843a981" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/fe/71/1c545fac6a9054b52b3771238fb2dc6e8f1d0ccec116e1c7786ec191887c/pytest_playwright-0.8.0-py3-none-any.whl", hash = "sha256:856aae6efd4bc055f2ef229c647768760bcaad5cd3a5983c314ac260a974a933" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "pytest-xdist"
version = "3.8.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "execnet" },
{ name = "pytest" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/78/b4/439b179d1ff526791eb921115fca8e44e596a13efeda518b9d845a619450/pytest_xdist-3.8.0.tar.gz", hash = "sha256:7e578125ec9bc6050861aa93f2d59f1d8d085595d6551c2c90b6f4fad8d3a9f1" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ca/31/d4e37e9e550c2b92a9cbc2e4d0b7420a27224968580b5a447f420847c975/pytest_xdist-3.8.0-py3-none-any.whl", hash = "sha256:202ca578cfeb7370784a8c33d6d05bc6e13b4f25b5053c30a152269fd10f0b88" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
[[package]]
name = "python-calamine"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "0.6.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/9b/32/99a794a1ca7b654cecdb76d4d61f21658b6f76574321341eb47df4365807/python_calamine-0.6.1.tar.gz", hash = "sha256:5974989919aa0bb55a136c1822d6f8b967d13c0fd0f245e3293abb4e63ab0f4b" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/f3/b6/d9b1a6432d33d43ded44ca01dff2c2a41f68a169413bdbe7677fc6598bfc/python_calamine-0.6.1-cp313-cp313-macosx_10_12_x86_64.whl", hash = "sha256:44dcffccbc3d9d258848d84ed685803ecb196f6b44bff271418283a0d015a6ea" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4d/09/29a113debc6c389065057c9f72e8837760b36ae86a6363a31c18b699adfb/python_calamine-0.6.1-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:66373ae273ef356a55b53c2348335274b6d25c08d75a399a3f167d93e13aa1b6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/89/c4/0a68314336b8b1d04ae1cda98cc8c191829547d652394f34e5360d9563c9/python_calamine-0.6.1-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:02482677cea6d3c2a09008469b7f5544d4d8c79af8fc7d49edcc669cfc75f640" },
{ url = "https://mirrors.aliyun.com/pypi/packages/29/ab/ce23029f808e31e12fe9ca26b038b67c8f065b9c666a1e73aacaa086d177/python_calamine-0.6.1-cp313-cp313-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:6794c55fa3d3dc88deda7377fc721b6506d186ec149e04b38109b1f58cc0b61f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/90/d9/e4bfad521a92ebb330f16a0ab7ad57da35ded14d90e9e395e97aacd63bef/python_calamine-0.6.1-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:79c33a5384221d8ab7d4b91b83374317b403ef945b5aa18f6e6ea6cbba661393" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ee/e8/18894883669644da9d14f8c6db0db00b793eaac3cd7268bcafb4a73b9837/python_calamine-0.6.1-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:0e36211a7feaa56d12e8ea1ddeeae6c4887764c351c275b034c07c9b7d66455e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0c/0d/7482fcded940d1adc4c8eaf47488a69ef1e3fd86eb8c6d33a981ddf5f82a/python_calamine-0.6.1-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:7c3e6aeedeb289311907f8d59b2a32a404433d1af4dfce0ba4e3badd30f9775d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ee/88/4898de6ce811c936168b48c92d310bba0e8f4ab6e56059b537d9d6d72c05/python_calamine-0.6.1-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:a2aa2989e5231cda2a15d21fd6e7cf3fc4ce09535756bdb7b2f32197fd6a566a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/10/1e/85ef4693452cc21cb912e32e33c8aa4add399b3fb0c1af8036692fd33f61/python_calamine-0.6.1-cp313-cp313-musllinux_1_1_aarch64.whl", hash = "sha256:e5761dc896446d6e9dd40c7e781908c1ae919d31bdd00b5dedc033525f440dec" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2f/18/67aaa61c4bea9fd99ed44ff50e93fac70096b992275bae3552f98f6a1229/python_calamine-0.6.1-cp313-cp313-musllinux_1_1_armv7l.whl", hash = "sha256:d1118d9d4f626f62663bfd5c83de07bc8455463081de6bc3b4264414e81a56a9" },
{ url = "https://mirrors.aliyun.com/pypi/packages/db/f5/73baef823b41f7b50a86ddb36d1ea2c19882414568aaa2d8ed7afb96dc71/python_calamine-0.6.1-cp313-cp313-musllinux_1_1_x86_64.whl", hash = "sha256:7e5500a8769bdf0efaef10bcce2613d5240823891172d1a943b776f18977c2f1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/aa/f2/db7fc4d14ff0bf8a8bf3ee43daad2e63fc2f46605e5972d97543e0f95e62/python_calamine-0.6.1-cp313-cp313-win32.whl", hash = "sha256:ec7928740519a8471ad8f1ec429301fb8a31a9c6adbfea51d7ff6ef2cb116835" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1d/c9/2e6b5d073885051ee7b5947156678c0cf5dfedf0dd10c5f23b694dcef824/python_calamine-0.6.1-cp313-cp313-win_amd64.whl", hash = "sha256:8f24740645a773cefae8507a13d03981867fa3dbd7fad1c3c667a1a3cd43235b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f2/c4/8ff9ecfe3b9b2bf556474e8ee8de541edfd650fd3e77752fa5705cbee3dc/python_calamine-0.6.1-cp313-cp313-win_arm64.whl", hash = "sha256:8e4ac2732aadc98bee412b59770dc6f4a6a886b5308cb57bfea53e877ae1a913" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ec/0d/83e44b3cbc7712ffac7750b14a817e34637904bcaa435626799506bf998b/python_calamine-0.6.1-cp313-cp313t-macosx_10_12_x86_64.whl", hash = "sha256:caab3bafa99b62d0aed0abf261a9f9df045eef11c5410ed91aa1b25f8381a087" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1f/7e/b47cfe737f885b139dae63f4139cb2ed1515994b465cf0370e25ce8d0065/python_calamine-0.6.1-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:3aefcdea5bdd2972e999264435b97e71855f02481688d213a4473d372b8288b0" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9a/ea/6aa2f277271323c5fbbde8718a7cad5ecf1fed9f637f648b0f6ae2c240cd/python_calamine-0.6.1-cp313-cp313t-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b2e9d10c91308eacfc1f76ff08bb7a8316c61f8f47619f9e4e254dd888fb3e9b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/00/2a/bf6ff24816fa60646d61a00f8a69113239a6a97207cdb2d541936003d030/python_calamine-0.6.1-cp313-cp313t-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:71eb5069b3d3639594a4fdccb3cb95a1b8f650e12def39a752ad8ff19eea620f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c1/24/54bb664dc9cc1252207bf5512d9870be23fdba2e5b94300d7e32e8c39a82/python_calamine-0.6.1-cp313-cp313t-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:def9e6be95496c660b6dc55b37eac3c6a479a71522e849f3a1ed4435788c6599" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c0/b7/4e2e5c8fd00ee7d80d272cb5e3cf170615a99911b515a2b4347995df0aa8/python_calamine-0.6.1-cp313-cp313t-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:c4740797c7e794acd907c7fa84ec09931ed2dfc3c9d1c689f7c7d236498d74cc" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b8/61/25193d600bf0e48513d275a69e5cdb158c27d11573bed74a28eb88d88592/python_calamine-0.6.1-cp313-cp313t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0b67f1a9f7452fa6ee736ac5a59349bbfc66087b96402051656c9b5a54a111ef" },
{ url = "https://mirrors.aliyun.com/pypi/packages/dc/3d/b0f434622c31182b64bd2e0e6c81cf35cf240ccee38cfb8074fbde9add98/python_calamine-0.6.1-cp313-cp313t-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:c1f369ebb8d6bf2ac66fbe38f5e6adf7b6a81fa71c1b5e2e7b2bb4a5c9667711" },
{ url = "https://mirrors.aliyun.com/pypi/packages/39/8e/502bbb06fa70f1f52f4f46efc0b331b31124110986a5378c1be711ad05e9/python_calamine-0.6.1-cp313-cp313t-musllinux_1_1_aarch64.whl", hash = "sha256:99bf12605466097219ebb133df54e41e479cb2559359d2dbad624dc301d4286b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c7/63/6fbda3f58aa5907cdfb628fc96e26e10820000858a9dd4fe6053e05a9310/python_calamine-0.6.1-cp313-cp313t-musllinux_1_1_armv7l.whl", hash = "sha256:96a44d48b9c4b05fb70396674ca7c90e4b4286845b5937606b60babe90f1fa4c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/07/e0/9e027e79de13424844ab33b6e2ad2b2be9ac40b653040bc8459bbfe4b48f/python_calamine-0.6.1-cp313-cp313t-musllinux_1_1_x86_64.whl", hash = "sha256:f7bfaf556391841ea59d0d0a63c5af7b5285ab260103656e65f55384b31b2010" },
{ url = "https://mirrors.aliyun.com/pypi/packages/cd/80/231c1f02d3d5adfde8c1f324da2c7907b63adb6f9ef36c3fd7db5b5fe083/python_calamine-0.6.1-cp313-cp313t-win_amd64.whl", hash = "sha256:a46ff8aa52ea3ef7851d6b5fd496be72a10db4f2d8942b42ecb0634ff9c1e441" },
{ url = "https://mirrors.aliyun.com/pypi/packages/88/2d/8c18519847dd53227c472231bcca37086027dd54b40ae13c48da7bacea53/python_calamine-0.6.1-cp313-cp313t-win_arm64.whl", hash = "sha256:7ac72743c3b2398ed55b9130482db097da8cb80d61b4b7aaf4008c7831ac11d3" },
Add fallback to use 'calamine' parse engine in excel_parser.py (#9374) ### What problem does this PR solve? add fallback to `calamine` engine when parse error raised using the default `openpyxl` / `xlrd` engine. e.g. the following error can be fixed: ``` Traceback (most recent call last): File "/ragflow/deepdoc/parser/excel_parser.py", line 53, in _load_excel_to_workbook df = pd.read_excel(file_like_object) File "/ragflow/.venv/lib/python3.10/site-packages/pandas/io/excel/_base.py", line 495, in read_excel io = ExcelFile( File "/ragflow/.venv/lib/python3.10/site-packages/pandas/io/excel/_base.py", line 1567, in __init__ self._reader = self._engines[engine]( File "/ragflow/.venv/lib/python3.10/site-packages/pandas/io/excel/_xlrd.py", line 46, in __init__ super().__init__( File "/ragflow/.venv/lib/python3.10/site-packages/pandas/io/excel/_base.py", line 573, in __init__ self.book = self.load_workbook(self.handles.handle, engine_kwargs) File "/ragflow/.venv/lib/python3.10/site-packages/pandas/io/excel/_xlrd.py", line 63, in load_workbook return open_workbook(file_contents=data, **engine_kwargs) File "/ragflow/.venv/lib/python3.10/site-packages/xlrd/__init__.py", line 172, in open_workbook bk = open_workbook_xls( File "/ragflow/.venv/lib/python3.10/site-packages/xlrd/book.py", line 68, in open_workbook_xls bk.biff2_8_load( File "/ragflow/.venv/lib/python3.10/site-packages/xlrd/book.py", line 641, in biff2_8_load cd.locate_named_stream(UNICODE_LITERAL(qname)) File "/ragflow/.venv/lib/python3.10/site-packages/xlrd/compdoc.py", line 398, in locate_named_stream result = self._locate_stream( File "/ragflow/.venv/lib/python3.10/site-packages/xlrd/compdoc.py", line 429, in _locate_stream raise CompDocError("%s corruption: seen[%d] == %d" % (qname, s, self.seen[s])) xlrd.compdoc.CompDocError: Workbook corruption: seen[2] == 4 ``` ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-08-12 12:41:33 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "python-dateutil"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "2.9.0.post0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "six" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/66/c0/0c8b6ad9f17a802ee498c46e004a0eb49bc148f2fd230864601a86dcf6db/python-dateutil-2.9.0.post0.tar.gz", hash = "sha256:37dd54208da7e1cd875388217d5e00ebd4179249f90fb72437e91a35459a0ad3" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ec/57/56b9bcc3c9c6a792fcbaf139543cee77261f3651ca9da0c93f5c1221264b/python_dateutil-2.9.0.post0-py2.py3-none-any.whl", hash = "sha256:a8b2bc7bffae282281c8140a97d3aa9c14da0b136dfe83f850eea9a5f7470427" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "python-docx"
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
version = "1.2.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "lxml" },
{ name = "typing-extensions" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a9/f7/eddfe33871520adab45aaa1a71f0402a2252050c14c7e3009446c8f4701c/python_docx-1.2.0.tar.gz", hash = "sha256:7bc9d7b7d8a69c9c02ca09216118c86552704edc23bac179283f2e38f86220ce" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/d0/00/1e03a4989fa5795da308cd774f05b704ace555a70f9bf9d3be057b680bcf/python_docx-1.2.0-py3-none-any.whl", hash = "sha256:3fd478f3250fbbbfd3b94fe1e985955737c145627498896a8a6bf81f4baf66c7" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "python-dotenv"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.2.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f0/26/19cadc79a718c5edbec86fd4919a6b6d3f681039a2f6d66d14be94e75fb9/python_dotenv-1.2.1.tar.gz", hash = "sha256:42667e897e16ab0d66954af0e60a9caa94f0fd4ecf3aaf6d2d260eec1aa36ad6" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/14/1b/a298b06749107c305e1fe0f814c6c74aea7b2f1e10989cb30f544a1b3253/python_dotenv-1.2.1-py3-none-any.whl", hash = "sha256:b81ee9561e9ca4004139c6cbba3a238c32b03e4894671e181b671e8cb8425d61" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: Gitlab connector (#12248) ### What problem does this PR solve? Feat: Gitlab connector Fix: submit button in darkmode ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2025-12-29 17:05:20 +08:00
[[package]]
name = "python-gitlab"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "7.1.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Gitlab connector (#12248) ### What problem does this PR solve? Feat: Gitlab connector Fix: submit button in darkmode ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2025-12-29 17:05:20 +08:00
dependencies = [
{ name = "requests" },
{ name = "requests-toolbelt" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/31/98/0b5d0a0367b90aec818298390b60ae65e6a08989cf5140271d0ee0206882/python_gitlab-7.1.0.tar.gz", hash = "sha256:1c34da3de40ad21675d788136f73d20a60649513e692f52c5a9720434db97c46" }
Feat: Gitlab connector (#12248) ### What problem does this PR solve? Feat: Gitlab connector Fix: submit button in darkmode ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2025-12-29 17:05:20 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/14/44/70fa1e395731b6a4b1f249d5f7326f3bb6281e2cf94d6535f679239f4b93/python_gitlab-7.1.0-py3-none-any.whl", hash = "sha256:8e42030cf27674e7ec9ea1f6d2fedcaaef0a6210f5fa22c80721abaa3a4fec90" },
Feat: Gitlab connector (#12248) ### What problem does this PR solve? Feat: Gitlab connector Fix: submit button in darkmode ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2025-12-29 17:05:20 +08:00
]
Refa: upgrade MCP SDK to v1.9.4 (#8421) ### What problem does this PR solve? Upgrade MCP SDK to v1.9.4 (latest). ### Type of change - [x] Refactoring
2025-06-23 16:53:59 +08:00
[[package]]
name = "python-multipart"
build(deps): bump python-multipart from 0.0.21 to 0.0.31 (#16088)
2026-06-17 19:39:42 +08:00
version = "0.0.31"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/64/7e/9b35ad8f3d9ca680f7c87a88f19612fdd8da9796c4d3b46e560ac79dcc4a/python_multipart-0.0.31.tar.gz", hash = "sha256:fc631183bb13e56db3158a4909908dfb2e23565286744e798241e63750e5d680" }
Refa: upgrade MCP SDK to v1.9.4 (#8421) ### What problem does this PR solve? Upgrade MCP SDK to v1.9.4 (latest). ### Type of change - [x] Refactoring
2025-06-23 16:53:59 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/5e/1e/7f7f299527a5a8ad90acd5f2f78dfa6c8495c6301a3205106ea68a84de96/python_multipart-0.0.31-py3-none-any.whl", hash = "sha256:8408153d68a9773291fc1da39a8b85a50044bddbabd2dd72e9229776b7b15e28" },
Refa: upgrade MCP SDK to v1.9.4 (#8421) ### What problem does this PR solve? Upgrade MCP SDK to v1.9.4 (latest). ### Type of change - [x] Refactoring
2025-06-23 16:53:59 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "python-pptx"
version = "1.0.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "lxml" },
{ name = "pillow" },
{ name = "typing-extensions" },
{ name = "xlsxwriter" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/52/a9/0c0db8d37b2b8a645666f7fd8accea4c6224e013c42b1d5c17c93590cd06/python_pptx-1.0.2.tar.gz", hash = "sha256:479a8af0eaf0f0d76b6f00b0887732874ad2e3188230315290cd1f9dd9cc7095" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/d9/4f/00be2196329ebbff56ce564aa94efb0fbc828d00de250b1980de1a34ab49/python_pptx-1.0.2-py3-none-any.whl", hash = "sha256:160838e0b8565a8b1f67947675886e9fea18aa5e795db7ae531606d68e785cba" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Update graspologic to gitee (#13362) ### What problem does this PR solve? Accelerate python module downloading ### Type of change - [x] Refactoring Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2026-03-04 17:48:47 +08:00
[[package]]
name = "python-slugify"
version = "8.0.4"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Update graspologic to gitee (#13362) ### What problem does this PR solve? Accelerate python module downloading ### Type of change - [x] Refactoring Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2026-03-04 17:48:47 +08:00
dependencies = [
{ name = "text-unidecode" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/87/c7/5e1547c44e31da50a460df93af11a535ace568ef89d7a811069ead340c4a/python-slugify-8.0.4.tar.gz", hash = "sha256:59202371d1d05b54a9e7720c5e038f928f45daaffe41dd10822f3907b937c856" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/a4/62/02da182e544a51a5c3ccf4b03ab79df279f9c60c5e82d5e8bec7ca26ac11/python_slugify-8.0.4-py2.py3-none-any.whl", hash = "sha256:276540b79961052b66b7d116620b36518847f52d5fd9e3a70164fc8c50faa6b8" },
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
]
[[package]]
name = "python-telegram-bot"
version = "22.7"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
dependencies = [
{ name = "httpx" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/e4/25/2258161b1069e66d6c39c0a602dbe57461d4767dc0012539970ea40bc9d6/python_telegram_bot-22.7.tar.gz", hash = "sha256:784b59ea3852fe4616ad63b4a0264c755637f5d725e87755ecdee28300febf61" }
Update graspologic to gitee (#13362) ### What problem does this PR solve? Accelerate python module downloading ### Type of change - [x] Refactoring Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2026-03-04 17:48:47 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/94/f7/0e2f89dd62f45d46d4ea0d8aec5893ce5b37389638db010c117f46f11450/python_telegram_bot-22.7-py3-none-any.whl", hash = "sha256:d72eed532cf763758cd9331b57a6d790aff0bb4d37d8f4e92149436fe21c6475" },
Update graspologic to gitee (#13362) ### What problem does this PR solve? Accelerate python module downloading ### Type of change - [x] Refactoring Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2026-03-04 17:48:47 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "pytz"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "2025.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f8/bf/abbd3cdfb8fbc7fb3d4d38d320f2441b1e7cbe29be4f23797b4a2b5d8aac/pytz-2025.2.tar.gz", hash = "sha256:360b9e3dbb49a209c21ad61809c7fb453643e048b38924c765813546746e81c3" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/81/c4/34e93fe5f5429d7570ec1fa436f1986fb1f00c3e0f43a589fe2bbcd22c3f/pytz-2025.2-py2.py3-none-any.whl", hash = "sha256:5ddf76296dd8c44c26eb8f4b6f35488f3ccbf6fbbd7adee0b7262d43f0ec2f00" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
[[package]]
name = "pywin32"
version = "311"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/a5/be/3fd5de0979fcb3994bfee0d65ed8ca9506a8a1260651b86174f6a86f52b3/pywin32-311-cp313-cp313-win32.whl", hash = "sha256:f95ba5a847cba10dd8c4d8fefa9f2a6cf283b8b88ed6178fa8a6c1ab16054d0d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e3/28/e0a1909523c6890208295a29e05c2adb2126364e289826c0a8bc7297bd5c/pywin32-311-cp313-cp313-win_amd64.whl", hash = "sha256:718a38f7e5b058e76aee1c56ddd06908116d35147e133427e59a3983f703a20d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/04/bf/90339ac0f55726dce7d794e6d79a18a91265bdf3aa70b6b9ca52f35e022a/pywin32-311-cp313-cp313-win_arm64.whl", hash = "sha256:7b4075d959648406202d92a2310cb990fea19b535c7f4a78d3f5e10b926eeb8a" },
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "pyyaml"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "6.0.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/05/8e/961c0007c59b8dd7729d542c61a4d537767a59645b82a0b521206e1e25c2/pyyaml-6.0.3.tar.gz", hash = "sha256:d76623373421df22fb4cf8817020cbb7ef15c725b9d5e45f17e189bfc384190f" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/d1/11/0fd08f8192109f7169db964b5707a2f1e8b745d4e239b784a5a1dd80d1db/pyyaml-6.0.3-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:8da9669d359f02c0b91ccc01cac4a67f16afec0dac22c2ad09f46bee0697eba8" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b1/16/95309993f1d3748cd644e02e38b75d50cbc0d9561d21f390a76242ce073f/pyyaml-6.0.3-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:2283a07e2c21a2aa78d9c4442724ec1eb15f5e42a723b99cb3d822d48f5f7ad1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/50/31/b20f376d3f810b9b2371e72ef5adb33879b25edb7a6d072cb7ca0c486398/pyyaml-6.0.3-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:ee2922902c45ae8ccada2c5b501ab86c36525b883eff4255313a253a3160861c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/49/1e/a55ca81e949270d5d4432fbbd19dfea5321eda7c41a849d443dc92fd1ff7/pyyaml-6.0.3-cp313-cp313-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:a33284e20b78bd4a18c8c2282d549d10bc8408a2a7ff57653c0cf0b9be0afce5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/74/27/e5b8f34d02d9995b80abcef563ea1f8b56d20134d8f4e5e81733b1feceb2/pyyaml-6.0.3-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:0f29edc409a6392443abf94b9cf89ce99889a1dd5376d94316ae5145dfedd5d6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f9/11/ba845c23988798f40e52ba45f34849aa8a1f2d4af4b798588010792ebad6/pyyaml-6.0.3-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:f7057c9a337546edc7973c0d3ba84ddcdf0daa14533c2065749c9075001090e6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3d/e0/7966e1a7bfc0a45bf0a7fb6b98ea03fc9b8d84fa7f2229e9659680b69ee3/pyyaml-6.0.3-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:eda16858a3cab07b80edaf74336ece1f986ba330fdb8ee0d6c0d68fe82bc96be" },
{ url = "https://mirrors.aliyun.com/pypi/packages/de/94/980b50a6531b3019e45ddeada0626d45fa85cbe22300844a7983285bed3b/pyyaml-6.0.3-cp313-cp313-win32.whl", hash = "sha256:d0eae10f8159e8fdad514efdc92d74fd8d682c933a6dd088030f3834bc8e6b26" },
{ url = "https://mirrors.aliyun.com/pypi/packages/97/c9/39d5b874e8b28845e4ec2202b5da735d0199dbe5b8fb85f91398814a9a46/pyyaml-6.0.3-cp313-cp313-win_amd64.whl", hash = "sha256:79005a0d97d5ddabfeeea4cf676af11e647e41d81c9a7722a193022accdb6b7c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/73/e8/2bdf3ca2090f68bb3d75b44da7bbc71843b19c9f2b9cb9b0f4ab7a5a4329/pyyaml-6.0.3-cp313-cp313-win_arm64.whl", hash = "sha256:5498cd1645aa724a7c71c8f378eb29ebe23da2fc0d7a08071d89469bf1d2defb" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "qianfan"
version = "0.4.6"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "aiohttp" },
{ name = "aiolimiter" },
{ name = "bce-python-sdk" },
{ name = "cachetools" },
{ name = "diskcache" },
{ name = "multiprocess" },
{ name = "prompt-toolkit" },
{ name = "pydantic" },
{ name = "python-dotenv" },
{ name = "pyyaml" },
{ name = "requests" },
{ name = "rich" },
{ name = "tenacity" },
{ name = "typer" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/3a/10/e06dd2f67a1f5f5a8eefac2e13c8f6d79c76e86a520c8976af33700d9c43/qianfan-0.4.6.tar.gz", hash = "sha256:90c2bf6f5fa1d1ae6ff63d982ce7b5fcb771c73048e81a147bcc24abed7eaefe" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/53/77/0fa3a283114078cd3b34465e4eeef03dae3cd94a63b81ec773d8883267a6/qianfan-0.4.6-py3-none-any.whl", hash = "sha256:7d8746356a2b88b42333e5fbb74c9ef7897d59b732e651a69e38e1be4512f0b1" },
Feat: Alter flask to Quart for async API serving. (#11275) ### What problem does this PR solve? #11277 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-18 17:05:16 +08:00
]
[[package]]
name = "quart"
version = "0.20.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Alter flask to Quart for async API serving. (#11275) ### What problem does this PR solve? #11277 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-18 17:05:16 +08:00
dependencies = [
{ name = "aiofiles" },
{ name = "blinker" },
{ name = "click" },
{ name = "flask" },
{ name = "hypercorn" },
{ name = "itsdangerous" },
{ name = "jinja2" },
{ name = "markupsafe" },
{ name = "werkzeug" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/1d/9d/12e1143a5bd2ccc05c293a6f5ae1df8fd94a8fc1440ecc6c344b2b30ce13/quart-0.20.0.tar.gz", hash = "sha256:08793c206ff832483586f5ae47018c7e40bdd75d886fee3fabbdaa70c2cf505d" }
Feat: Alter flask to Quart for async API serving. (#11275) ### What problem does this PR solve? #11277 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-18 17:05:16 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/7e/e9/cc28f21f52913adf333f653b9e0a3bf9cb223f5083a26422968ba73edd8d/quart-0.20.0-py3-none-any.whl", hash = "sha256:003c08f551746710acb757de49d9b768986fd431517d0eb127380b656b98b8f1" },
Feat: Alter flask to Quart for async API serving. (#11275) ### What problem does this PR solve? #11277 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-18 17:05:16 +08:00
]
[[package]]
name = "quart-auth"
version = "0.11.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Alter flask to Quart for async API serving. (#11275) ### What problem does this PR solve? #11277 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-18 17:05:16 +08:00
dependencies = [
{ name = "quart" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/45/8d/d965905118fe612e7520f8a1014e4988842d8ed9fd68de4471999f7d968b/quart_auth-0.11.0.tar.gz", hash = "sha256:7703df693d795b3ec43a634efe4118c6adbddd98e7b8195008ca6def6d45cb47" }
Feat: Alter flask to Quart for async API serving. (#11275) ### What problem does this PR solve? #11277 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-18 17:05:16 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/cf/a3/8ca9235569f39e471a9a319ba1665a27cb9cea8c48c688966460db67d48f/quart_auth-0.11.0-py3-none-any.whl", hash = "sha256:dd342ea39475a9b32b79d83e2b6820ddaa358e77f01dedbba47d50529f2c8c74" },
Feat: Alter flask to Quart for async API serving. (#11275) ### What problem does this PR solve? #11277 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-18 17:05:16 +08:00
]
[[package]]
name = "quart-cors"
version = "0.8.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Alter flask to Quart for async API serving. (#11275) ### What problem does this PR solve? #11277 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-18 17:05:16 +08:00
dependencies = [
{ name = "quart" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/14/b1/2a65be601f3c92c913f3321ee186d10c2da4325447b4b0fca83e0c493c60/quart_cors-0.8.0.tar.gz", hash = "sha256:ac32c4931da6fba944e9e2d3f856f2db4fd82e3fb905a09646086780c221a118" }
Feat: Alter flask to Quart for async API serving. (#11275) ### What problem does this PR solve? #11277 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-18 17:05:16 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ea/31/da390a5a10674481dea2909178973de81fa3a246c0eedcc0e1e4114f52f8/quart_cors-0.8.0-py3-none-any.whl", hash = "sha256:62dc811768e2e1704d2b99d5880e3eb26fc776832305a19ea53db66f63837767" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: support openapi (#12521) ### What problem does this PR solve? Support OpenAPI interface description. The issue of not supporting the Swagger interface after upgrading the system framework from Flask to Quart has been resolved. Resolved https://github.com/infiniflow/ragflow/issues/5264 ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: puhaiyang <“761396462@qq.com”>
2026-01-09 17:48:20 +08:00
[[package]]
name = "quart-schema"
version = "0.23.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: support openapi (#12521) ### What problem does this PR solve? Support OpenAPI interface description. The issue of not supporting the Swagger interface after upgrading the system framework from Flask to Quart has been resolved. Resolved https://github.com/infiniflow/ragflow/issues/5264 ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: puhaiyang <“761396462@qq.com”>
2026-01-09 17:48:20 +08:00
dependencies = [
{ name = "pyhumps" },
{ name = "quart" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/24/65/97b69c76bc8838f0389387c87f480382eea48ca60d5262aeaf4086ad14e2/quart_schema-0.23.0.tar.gz", hash = "sha256:778f36aa80697420a0148807eb324b7d6ca1f10793cd1d0eb4f1c7908d860bdd" }
Feat: support openapi (#12521) ### What problem does this PR solve? Support OpenAPI interface description. The issue of not supporting the Swagger interface after upgrading the system framework from Flask to Quart has been resolved. Resolved https://github.com/infiniflow/ragflow/issues/5264 ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: puhaiyang <“761396462@qq.com”>
2026-01-09 17:48:20 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/12/ba/54c4516499bf6549ff47d656b8dc8cd58cea7f6d03d3097aebf1958f4974/quart_schema-0.23.0-py3-none-any.whl", hash = "sha256:f8f217942d433954dfe9860b4d748fe4b111836d8d74e06bc0afc512dd991c80" },
Feat: support openapi (#12521) ### What problem does this PR solve? Support OpenAPI interface description. The issue of not supporting the Swagger interface after upgrading the system framework from Flask to Quart has been resolved. Resolved https://github.com/infiniflow/ragflow/issues/5264 ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: puhaiyang <“761396462@qq.com”>
2026-01-09 17:48:20 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "ragflow"
Docs: Update version references to v0.26.2 in READMEs and docs (#16387)
2026-06-29 09:40:45 +08:00
version = "0.26.2"
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
source = { virtual = "." }
dependencies = [
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
{ name = "agentrun-sdk" },
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
{ name = "aiohttp" },
Feat: Add box connector (#11845) ### What problem does this PR solve? Feat: Add box connector ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-12 10:23:40 +08:00
{ name = "aiosmtplib" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "akshare" },
Feat: add DingTalk AI Table connector and integration for data synch… (#13413) ### What problem does this PR solve? Add DingTalk AI Table connector and integration for data synchronization Issue #13400 ### Type of change - [x] New Feature (non-breaking change which adds functionality) Co-authored-by: wangheyang <wangheyang@corp.netease.com>
2026-03-06 21:13:23 +08:00
{ name = "alibabacloud-dingtalk" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "anthropic" },
{ name = "arxiv" },
Feat: Add Asana data source integration and configuration options (#12239) ### What problem does this PR solve? change: Add Asana data source integration and configuration options ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2025-12-29 13:28:37 +08:00
{ name = "asana" },
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
{ name = "atlassian-python-api" },
feat: bump Python minimum from 3.12 to 3.13, drop strenum backport (#14767) Closes #14753 ## What changed | File | Change | |---|---| | `pyproject.toml` | `requires-python` → `>=3.13,<3.15`; remove `strenum==0.4.15` | | `Dockerfile` | `uv python install 3.13`, `uv sync --python 3.13` | | `.github/workflows/tests.yml` | `uv sync --python 3.13` on both matrix legs | | `CLAUDE.md` | dev setup command + requirements note updated | | `deepdoc/parser/mineru_parser.py` | `from strenum import StrEnum` → `from enum import StrEnum` | | `agent/tools/code_exec.py` | same | `StrEnum` has been in the stdlib since Python 3.11 — the `strenum` backport package is no longer needed once the floor is 3.13. ## Why uv.lock is not regenerated `uv lock --python 3.13` fails because: 1. The infiniflow/graspologic fork pins `numpy>=1.26.4,<2.0.0` 2. `tensorflow-cpu>=2.20.0` (the first release with cp313 wheels) depends on `ml-dtypes>=0.5.1`, which requires `numpy>=2.1.0` 3. These two constraints are irreconcilable on Python 3.13 The lockfile regeneration requires loosening the `numpy` upper bound in the `infiniflow/graspologic` fork. Once that fork commit is updated and the SHA in `pyproject.toml:49` is bumped, `uv lock --python 3.13` will succeed. ## RFC corrections Two claims in the original RFC (#14753) did not hold up under code review: - **"graspologic hard-blocks 3.13"** — the infiniflow fork at the pinned commit has no `<3.13` Python constraint. The blocker is the transitive `numpy<2.0.0` conflict with tensorflow-cpu's test dependency, not a direct Python version cap. - **"free-threading throughput gains for I/O-bound workload"** — Python 3.13 free-threading requires a special `--disable-gil` build and provides no benefit for async I/O code (the GIL is already released during I/O). The real motivation is forward compatibility and improved error messages.
2026-05-15 08:40:53 +02:00
{ name = "audioop-lts" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "azure-identity" },
{ name = "azure-storage-file-datalake" },
{ name = "beartype" },
{ name = "bio" },
Feat: add BedrockCV for vision/image2text inference via LiteLLM (#14705) ## Summary - `CvModel["Bedrock"]` was absent from `rag/llm/cv_model.py`, causing `model_instance()` to return `None` when a Bedrock model was used as a PDF parser — even after correct model resolution. - This PR adds `BedrockCV`, enabling Bedrock vision models (e.g. `amazon.nova-pro-v1:0`, `anthropic.claude-3-5-sonnet`) to be used as PDF parsers. ## What problem does this PR solve? When a Bedrock model is selected as the PDF parser in a knowledge base, ingestion failed with: ``` 'LiteLLMBase' object has no attribute 'describe_with_prompt' ``` The root cause: `LiteLLMBase` (the Bedrock chat implementation) was the only registered handler for the Bedrock factory. It does not implement `describe_with_prompt`. `CvModel` had no Bedrock entry, so `model_instance()` returned `None` for `image2text` requests. ## Type of change - [x] New Feature (non-breaking change which adds functionality) ## Changes **`rag/llm/cv_model.py`** Adds `BedrockCV(Base)` with `_FACTORY_NAME = "Bedrock"`: - Uses `litellm.completion` with the `bedrock/` prefix (consistent with `LiteLLMBase`) - Parses AWS credentials from the JSON key assembled by `add_llm` (`auth_mode`, `bedrock_ak`, `bedrock_sk`, `bedrock_region`, `aws_role_arn`) - Supports three auth modes: `access_key_secret`, `iam_role` (via STS `assume_role`), and default credential chain (IRSA, instance profile) - Implements `describe_with_prompt` and `describe` ## Test plan - [ ] Configure a Bedrock vision model (e.g. `amazon.nova-pro-v1:0`) with valid AWS credentials - [ ] Select it as PDF parser in a knowledge base - [ ] Verify ingestion of a PDF document completes without errors - [ ] Verify `CvModel["Bedrock"]` resolves to `BedrockCV` 🤖 Generated with [Claude Code](https://claude.ai/claude-code) --------- Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-05-11 04:29:58 +02:00
{ name = "boto3" },
Feat: Add box connector (#11845) ### What problem does this PR solve? Feat: Add box connector ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-12 10:23:40 +08:00
{ name = "boxsdk" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
{ name = "browser-use" },
Feat: add forgot password reset, solve #8547 (#10586) ### What problem does this PR solve? Feat: add forgot password reset, solve #8547 ### Type of change - [X] New Feature (non-breaking change which adds functionality)
2025-10-16 15:07:49 +08:00
{ name = "captcha" },
Feat: optimize ingestion pipeline with preprocess (#13211) ### What problem does this PR solve? Feat: optimize ingestion pipeline with preprocess ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-02-26 10:24:13 +08:00
{ name = "chardet" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "cn2an" },
{ name = "cohere" },
{ name = "crawl4ai" },
{ name = "dashscope" },
Go: add balance command (#14262) ### What problem does this PR solve? ``` RAGFlow(user)> list supported models from 'moonshot' 'test'; +---------------------------------+ | model_name | +---------------------------------+ | moonshot-v1-32k-vision-preview | | kimi-k2.6 | | moonshot-v1-8k | | moonshot-v1-auto | | moonshot-v1-128k | | moonshot-v1-32k | | kimi-k2.5 | | moonshot-v1-8k-vision-preview | | moonshot-v1-128k-vision-preview | +---------------------------------+ RAGFlow(user)> show balance from 'moonshot' 'test'; +---------+----------+ | balance | currency | +---------+----------+ | 0 | CNY | +---------+----------+ ``` ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2026-04-21 21:31:50 +08:00
{ name = "debugpy" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "deepl" },
{ name = "demjson3" },
{ name = "discord-py" },
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
{ name = "dropbox" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "duckduckgo-search" },
{ name = "editdistance" },
{ name = "elasticsearch-dsl" },
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
{ name = "en-core-web-sm" },
Use Infinity single-field-multi-index (#11444) ### What problem does this PR solve? Use Infinity single-field-multi-index ### Type of change - [x] Refactoring - [x] Performance Improvement
2025-11-26 11:06:37 +08:00
{ name = "exceptiongroup" },
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
{ name = "extract-msg" },
feat: support rss datasource (#13721) ### What problem does this PR solve? Supporting public RSS/Atom feed URLs as data sources for RagFlow. link https://github.com/infiniflow/ragflow/issues/12313 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-03-27 22:58:44 +08:00
{ name = "feedparser" },
Feat:new api /sequence2txt and update QWenSeq2txt (#11643) ### What problem does this PR solve? change: new api /sequence2txt, update QWenSeq2txt and ZhipuSeq2txt ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-02 11:17:31 +08:00
{ name = "ffmpeg-python" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "flasgger" },
{ name = "flask-cors" },
{ name = "flask-login" },
Feat: add SMTP support for user invitation emails (#9479) ### What problem does this PR solve? Add SMTP support for user invitation emails ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-08-15 18:12:20 +08:00
{ name = "flask-mail" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "flask-session" },
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
{ name = "google-api-python-client" },
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
{ name = "google-auth-oauthlib" },
build(deps): bump crawl4ai from 0.8.9 to 0.9.0 (#16470)
2026-06-30 09:34:48 +08:00
{ name = "google-cloud-bigquery" },
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
{ name = "google-cloud-storage" },
Fix problem with Google Cloud models with reasoning (like gemini) - Additional fix to issue #10474 (#10502) ### What problem does this PR solve? Issue #10474 - Update to PR #10477 ### Type of change - [X] Bug Fix (non-breaking change which fixes an issue)
2025-10-15 08:54:20 +02:00
{ name = "google-genai" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "google-search-results" },
{ name = "graspologic" },
{ name = "groq" },
Remove unused py module dependencies (#11964) ### What problem does this PR solve? ``` pip list --not-required Package Version ---------------------------- --------------------- aiosmtplib 5.0.0 akshare 1.17.94 anthropic 0.34.1 arxiv 2.1.3 Aspose.Slides 24.7.0 atlassian-python-api 4.0.7 azure-identity 1.17.1 azure-storage-file-datalake 12.16.0 bio 1.7.1 boxsdk 10.2.0 captcha 0.7.1 cn2an 0.5.22 cohere 5.6.2 Crawl4AI 0.4.247 dashscope 1.20.11 deepl 1.18.0 demjson3 3.0.6 discord.py 2.3.2 dropbox 12.0.2 duckduckgo_search 7.5.5 editdistance 0.8.1 elasticsearch-dsl 8.12.0 exceptiongroup 1.3.1 extract-msg 0.55.0 ffmpeg-python 0.2.0 flasgger 0.9.7.1 Flask-Cors 5.0.0 Flask-Login 0.6.3 Flask-Mail 0.10.0 Flask-Session 0.8.0 google-auth-oauthlib 1.2.3 google-genai 1.55.0 google-generativeai 0.8.5 google_search_results 2.4.2 graspologic 0.1.dev847+g38e680cab groq 0.9.0 grpcio-status 1.67.1 html_text 0.6.2 imageio-ffmpeg 0.6.0 infinity_emb 0.0.66 infinity-sdk 0.6.11 jira 3.10.5 json_repair 0.35.0 langfuse 3.10.5 mammoth 1.11.0 Markdown 3.6 markdown_to_json 2.1.1 markdownify 1.2.2 mcp 1.19.0 mini-racer 0.12.4 minio 7.2.4 mistralai 0.4.2 moodlepy 0.24.1 mypy-boto3-s3 1.40.26 Office365-REST-Python-Client 2.6.2 ollama 0.6.1 onnxruntime-gpu 1.23.2 opencv-python 4.10.0.84 opencv-python-headless 4.10.0.84 opendal 0.45.20 opensearch-py 2.7.1 ormsgpack 1.5.0 pdfplumber 0.10.4 pip 25.3 pluginlib 0.9.4 psycopg2-binary 2.9.11 pyclipper 1.4.0 pycryptodomex 3.20.0 pyobvector 0.2.18 pyodbc 5.3.0 pypandoc 1.16.2 pypdf 6.4.0 PyPDF2 3.0.1 python-calamine 0.6.1 python-docx 1.2.0 python-pptx 1.0.2 pywencai 0.13.1 qianfan 0.4.6 quart-auth 0.11.0 quart-cors 0.8.0 ranx 0.3.20 readability-lxml 0.8.4.1 replicate 0.31.0 reportlab 4.4.6 roman-numbers 1.0.2 ruamel.base 1.0.0 ruamel.yaml 0.18.16 scholarly 1.7.11 selenium-wire 5.1.0 slack_sdk 3.37.0 socksio 1.0.0 sqlglotrs 0.9.0 StrEnum 0.4.15 tavily-python 0.5.1 tencentcloud-sdk-python 3.0.1478 tika 2.6.0 valkey 6.0.2 vertexai 1.70.0 volcengine 1.0.194 voyageai 0.2.3 webdav4 0.10.0 webdriver-manager 4.0.1 wikipedia 1.4.0 word2number 1.1 xgboost 1.6.0 xpinyin 0.7.6 yfinance 0.2.65 zhipuai 2.0.1 ``` ### Type of change - [x] Refactoring Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-16 12:40:03 +08:00
{ name = "grpcio-status" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "html-text" },
{ name = "infinity-emb" },
{ name = "infinity-sdk" },
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
{ name = "jira" },
Light GraphRAG (#4585) ### What problem does this PR solve? #4543 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-01-22 19:43:14 +08:00
{ name = "json-repair" },
Feat: add basic Langfuse support for LLM module (#6443) ### What problem does this PR solve? #6155 Add basic Langfuse support for LLM module. A trace example: <img width="755" alt="image" src="https://github.com/user-attachments/assets/25c1f852-5116-486c-a47f-6097187142ca" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-03-24 13:18:47 +08:00
{ name = "langfuse" },
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
{ name = "lark-oapi" },
{ name = "line-bot-sdk" },
Security: do not use litellm 1.82.7 and 1.82.8 (#13768) ### What problem does this PR solve? See [issue](https://github.com/BerriAI/litellm/issues/24518) from Litellm. Upgraded from `1.81.15` to `1.82.6`, so RAGFlow is safe as always. ### Type of change - [x] Security Co-authored-by: Jin Hai <haijin.chn@gmail.com> Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
2026-03-25 22:39:33 +08:00
{ name = "litellm" },
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
{ name = "mammoth" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "markdown" },
{ name = "markdown-to-json" },
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
{ name = "markdownify" },
Feat: mcp server (#7084) ### What problem does this PR solve? Add MCP support with a client example. Issue link: #4344 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-04-21 09:43:20 +08:00
{ name = "mcp" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "mini-racer" },
{ name = "minio" },
{ name = "mistralai" },
Add Moodle data source integration (#11325) ### What problem does this PR solve? This PR adds a native Moodle connector to sync content (courses, resources, forums, assignments, pages, books) into RAGFlow. ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2025-11-21 12:58:49 +01:00
{ name = "moodlepy" },
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
{ name = "mypy-boto3-s3" },
feat/add MySQL and PostgreSQL data source connectors (#12817) ### What problem does this PR solve? This PR adds MySQL and PostgreSQL as data source connectors, allowing users to import data directly from relational databases into RAGFlow for RAG workflows. Many users store their knowledge in databases (product catalogs, documentation, FAQs, etc.) and currently have no way to sync this data into RAGFlow without exporting to files first. This feature lets them connect directly to their databases, run SQL queries, and automatically create documents from the results. Closes #763 Closes #11560 ### Type of change - [ ] Bug Fix (non-breaking change which fixes an issue) - [x] New Feature (non-breaking change which adds functionality) - [ ] Documentation Update - [ ] Refactoring - [ ] Performance Improvement - [ ] Other (please describe): ### What this PR does **New capabilities:** - Connect to MySQL and PostgreSQL databases - Run custom SQL queries to extract data - Map database columns to document content (vectorized) and metadata (searchable) - Support incremental sync using a timestamp column - Full frontend UI with connection form and tooltips **Files changed:** Backend: - `common/constants.py` - Added MYSQL/POSTGRESQL to FileSource enum - `common/data_source/config.py` - Added to DocumentSource enum - `common/data_source/rdbms_connector.py` - New connector (368 lines) - `common/data_source/__init__.py` - Exported the connector - `rag/svr/sync_data_source.py` - Added MySQL and PostgreSQL sync classes - `pyproject.toml` - Added mysql-connector-python dependency Frontend: - `web/src/pages/user-setting/data-source/constant/index.tsx` - Form fields - `web/src/locales/en.ts` - English translations - `web/src/assets/svg/data-source/mysql.svg` - MySQL icon - `web/src/assets/svg/data-source/postgresql.svg` - PostgreSQL icon ### Testing done Tested with MySQL 8.0 and PostgreSQL 16: - Connection validation works correctly - Full sync imports all query results as documents - Incremental sync only fetches rows updated since last sync - Custom SQL queries filter data as expected - Invalid credentials show clear error messages - Lint checks pass (`ruff check` returns no errors) --------- Co-authored-by: mkdev11 <YOUR_GITHUB_ID+MkDev11@users.noreply.github.com>
2026-02-03 23:14:32 -03:00
{ name = "mysql-connector-python" },
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
{ name = "nest-asyncio" },
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
{ name = "office365-rest-python-client" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "ollama" },
Fix onnxruntime-gpu marks (#4643) ### What problem does this PR solve? Fix onnxruntime-gpu marks ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-01-26 09:37:59 +08:00
{ name = "onnxruntime", marker = "platform_machine != 'x86_64' or sys_platform == 'darwin'" },
{ name = "onnxruntime-gpu", marker = "platform_machine == 'x86_64' and sys_platform != 'darwin'" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "opencv-python" },
{ name = "opencv-python-headless" },
Oss support opendal(including mysql) (#8204) ### What problem does this PR solve? #8074 Oss support opendal(including mysql) ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2025-06-12 11:37:42 +08:00
{ name = "opendal" },
Feat: Adds OpenSearch2.19.1 as the vector_database support (#7140) ### What problem does this PR solve? This PR adds the support for latest OpenSearch2.19.1 as the store engine & search engine option for RAGFlow. ### Main Benefit 1. OpenSearch2.19.1 is licensed under the [Apache v2.0 License] which is much better than Elasticsearch 2. For search, OpenSearch2.19.1 supports full-text search、vector_search、hybrid_search those are similar with Elasticsearch on schema 3. For store, OpenSearch2.19.1 stores text、vector those are quite simliar with Elasticsearch on schema ### Changes - Support opensearch_python_connetor. I make a lot of adaptions since the schema and api/method between ES and Opensearch differs in many ways(especially the knn_search has a significant gap) : rag/utils/opensearch_coon.py - Support static config adaptions by changing: conf/service_conf.yaml、api/settings.py、rag/settings.py - Supprt some store&search schema changes between OpenSearch and ES: conf/os_mapping.json - Support OpenSearch python sdk : pyproject.toml - Support docker config for OpenSearch2.19.1 : docker/.env、docker/docker-compose-base.yml、docker/service_conf.yaml.template ### How to use - I didn't change the priority that ES as the default doc/search engine. Only if in docker/.env , we set DOC_ENGINE=${DOC_ENGINE:-opensearch}, it will work. ### Others Our team tested a lot of docs in our environment by using OpenSearch as the vector database ,it works very well. All the conifg for OpenSearch is necessary. ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Yongteng Lei <yongtengrey@outlook.com> Co-authored-by: writinwaters <93570324+writinwaters@users.noreply.github.com> Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
2025-04-24 16:03:31 +08:00
{ name = "opensearch-py" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "ormsgpack" },
Feat: add local & ssh provider in admin panel (#15039) ### What problem does this PR solve? Feat: add local & ssh provider in admin panel ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-20 16:56:20 +08:00
{ name = "paramiko" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "pdfplumber" },
Feat: optimize ingestion pipeline with preprocess (#13211) ### What problem does this PR solve? Feat: optimize ingestion pipeline with preprocess ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-02-26 10:24:13 +08:00
{ name = "peewee" },
Feat: Support tool calling in Generate component (#7572) ### What problem does this PR solve? Hello, our use case requires LLM agent to invoke some tools, so I made a simple implementation here. This PR does two things: 1. A simple plugin mechanism based on `pluginlib`: This mechanism lives in the `plugin` directory. It will only load plugins from `plugin/embedded_plugins` for now. A sample plugin `bad_calculator.py` is placed in `plugin/embedded_plugins/llm_tools`, it accepts two numbers `a` and `b`, then give a wrong result `a + b + 100`. In the future, it can load plugins from external location with little code change. Plugins are divided into different types. The only plugin type supported in this PR is `llm_tools`, which must implement the `LLMToolPlugin` class in the `plugin/llm_tool_plugin.py`. More plugin types can be added in the future. 2. A tool selector in the `Generate` component: Added a tool selector to select one or more tools for LLM: ![image](https://github.com/user-attachments/assets/74a21fdf-9333-4175-991b-43df6524c5dc) And with the `bad_calculator` tool, it results this with the `qwen-max` model: ![image](https://github.com/user-attachments/assets/93aff9c4-8550-414a-90a2-1a15a5249d94) ### Type of change - [ ] Bug Fix (non-breaking change which fixes an issue) - [x] New Feature (non-breaking change which adds functionality) - [ ] Documentation Update - [ ] Refactoring - [ ] Performance Improvement - [ ] Other (please describe): Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
2025-05-16 16:32:19 +08:00
{ name = "pluginlib" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "psycopg2-binary" },
Feat: add Airtable connector and integration for data synchronization (#12211) ### What problem does this PR solve? change: add Airtable connector and integration for data synchronization ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-25 17:50:41 +08:00
{ name = "pyairtable" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "pyclipper" },
{ name = "pycryptodomex" },
Feat: GitHub connector (#12314) ### What problem does this PR solve? Feat: GitHub connector ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-30 15:09:52 +08:00
{ name = "pygithub" },
feat: add OceanBase doc engine (#11228) ### What problem does this PR solve? Add OceanBase doc engine. Close #5350 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-20 10:00:14 +08:00
{ name = "pyobvector" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "pyodbc" },
Feat: extract message output to file (#11251) ### What problem does this PR solve? Feat: extract message output to file ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-14 19:52:11 +08:00
{ name = "pypandoc" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "pypdf" },
Add fallback to use 'calamine' parse engine in excel_parser.py (#9374) ### What problem does this PR solve? add fallback to `calamine` engine when parse error raised using the default `openpyxl` / `xlrd` engine. e.g. the following error can be fixed: ``` Traceback (most recent call last): File "/ragflow/deepdoc/parser/excel_parser.py", line 53, in _load_excel_to_workbook df = pd.read_excel(file_like_object) File "/ragflow/.venv/lib/python3.10/site-packages/pandas/io/excel/_base.py", line 495, in read_excel io = ExcelFile( File "/ragflow/.venv/lib/python3.10/site-packages/pandas/io/excel/_base.py", line 1567, in __init__ self._reader = self._engines[engine]( File "/ragflow/.venv/lib/python3.10/site-packages/pandas/io/excel/_xlrd.py", line 46, in __init__ super().__init__( File "/ragflow/.venv/lib/python3.10/site-packages/pandas/io/excel/_base.py", line 573, in __init__ self.book = self.load_workbook(self.handles.handle, engine_kwargs) File "/ragflow/.venv/lib/python3.10/site-packages/pandas/io/excel/_xlrd.py", line 63, in load_workbook return open_workbook(file_contents=data, **engine_kwargs) File "/ragflow/.venv/lib/python3.10/site-packages/xlrd/__init__.py", line 172, in open_workbook bk = open_workbook_xls( File "/ragflow/.venv/lib/python3.10/site-packages/xlrd/book.py", line 68, in open_workbook_xls bk.biff2_8_load( File "/ragflow/.venv/lib/python3.10/site-packages/xlrd/book.py", line 641, in biff2_8_load cd.locate_named_stream(UNICODE_LITERAL(qname)) File "/ragflow/.venv/lib/python3.10/site-packages/xlrd/compdoc.py", line 398, in locate_named_stream result = self._locate_stream( File "/ragflow/.venv/lib/python3.10/site-packages/xlrd/compdoc.py", line 429, in _locate_stream raise CompDocError("%s corruption: seen[%d] == %d" % (qname, s, self.seen[s])) xlrd.compdoc.CompDocError: Workbook corruption: seen[2] == 4 ``` ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-08-12 12:41:33 +08:00
{ name = "python-calamine" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "python-docx" },
Feat: Gitlab connector (#12248) ### What problem does this PR solve? Feat: Gitlab connector Fix: submit button in darkmode ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2025-12-29 17:05:20 +08:00
{ name = "python-gitlab" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "python-pptx" },
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
{ name = "python-telegram-bot" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "qianfan" },
Feat: Alter flask to Quart for async API serving. (#11275) ### What problem does this PR solve? #11277 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-18 17:05:16 +08:00
{ name = "quart-auth" },
{ name = "quart-cors" },
Feat: support openapi (#12521) ### What problem does this PR solve? Support OpenAPI interface description. The issue of not supporting the Swagger interface after upgrading the system framework from Flask to Quart has been resolved. Resolved https://github.com/infiniflow/ragflow/issues/5264 ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: puhaiyang <“761396462@qq.com”>
2026-01-09 17:48:20 +08:00
{ name = "quart-schema" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "ranx" },
{ name = "readability-lxml" },
{ name = "replicate" },
Feature/docs generator (#11858) ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### What problem does this PR solve? This PR introduces a new Docs Generator agent component for producing downloadable PDF, DOCX, or TXT files from Markdown content generated within a RAGFlow workflow. ### **Key Features** **Backend** - New component: DocsGenerator (agent/component/docs_generator.py) - - Markdown → PDF/DOCX/TXT conversion - - Supports tables, lists, code blocks, headings, and rich formatting - - Configurable document style (fonts, margins, colors, page size, orientation) - - Optional header logo and footer with page numbers/timestamps - **Frontend** - New configuration UI for the Docs Generator - - Download button integrated into the chat interface - - Output wired to the Message component - - Full i18n support **Documentation** Added component guide: docs/guides/agent/agent_component_reference/docs_generator.md **Usage** Add the Docs Generator to a workflow, connect Markdown output from an upstream component, configure metadata/style, and feed its output into the Message component. Users will see a document download button directly in the chat. **Contributor Note** We have been following RAGFlow since more than a year and half now and have worked extensively on personalizing the framework and integrating it into several of our internal systems. Over the past year and a half, we have built multiple platforms that rely on RAGFlow as a core component, which has given us a strong appreciation for how flexible and powerful the project is. We also previously contributed the full Italian translation, and we were glad to see it accepted. This new Docs Generator component was created for our own production needs, and we believe that it may be useful for many others in the community as well. We want to sincerely thank the entire RAGFlow team for the remarkable work you have done and continue to do. If there are opportunities to contribute further, we would be glad to help whenever we have time available. It would be a pleasure to support the project in any way we can. If appropriate, we would be glad to be listed among the project’s contributors, but in any case we look forward to continuing to support and contribute to the project. PentaFrame Development Team --------- Co-authored-by: PentaFrame <info@pentaframe.it> Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2025-12-12 07:59:43 +01:00
{ name = "reportlab" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "roman-numbers" },
{ name = "ruamel-base" },
{ name = "ruamel-yaml" },
{ name = "scholarly" },
{ name = "selenium-wire" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
{ name = "setuptools" },
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
{ name = "slack-sdk" },
Remove unused py module dependencies (#11964) ### What problem does this PR solve? ``` pip list --not-required Package Version ---------------------------- --------------------- aiosmtplib 5.0.0 akshare 1.17.94 anthropic 0.34.1 arxiv 2.1.3 Aspose.Slides 24.7.0 atlassian-python-api 4.0.7 azure-identity 1.17.1 azure-storage-file-datalake 12.16.0 bio 1.7.1 boxsdk 10.2.0 captcha 0.7.1 cn2an 0.5.22 cohere 5.6.2 Crawl4AI 0.4.247 dashscope 1.20.11 deepl 1.18.0 demjson3 3.0.6 discord.py 2.3.2 dropbox 12.0.2 duckduckgo_search 7.5.5 editdistance 0.8.1 elasticsearch-dsl 8.12.0 exceptiongroup 1.3.1 extract-msg 0.55.0 ffmpeg-python 0.2.0 flasgger 0.9.7.1 Flask-Cors 5.0.0 Flask-Login 0.6.3 Flask-Mail 0.10.0 Flask-Session 0.8.0 google-auth-oauthlib 1.2.3 google-genai 1.55.0 google-generativeai 0.8.5 google_search_results 2.4.2 graspologic 0.1.dev847+g38e680cab groq 0.9.0 grpcio-status 1.67.1 html_text 0.6.2 imageio-ffmpeg 0.6.0 infinity_emb 0.0.66 infinity-sdk 0.6.11 jira 3.10.5 json_repair 0.35.0 langfuse 3.10.5 mammoth 1.11.0 Markdown 3.6 markdown_to_json 2.1.1 markdownify 1.2.2 mcp 1.19.0 mini-racer 0.12.4 minio 7.2.4 mistralai 0.4.2 moodlepy 0.24.1 mypy-boto3-s3 1.40.26 Office365-REST-Python-Client 2.6.2 ollama 0.6.1 onnxruntime-gpu 1.23.2 opencv-python 4.10.0.84 opencv-python-headless 4.10.0.84 opendal 0.45.20 opensearch-py 2.7.1 ormsgpack 1.5.0 pdfplumber 0.10.4 pip 25.3 pluginlib 0.9.4 psycopg2-binary 2.9.11 pyclipper 1.4.0 pycryptodomex 3.20.0 pyobvector 0.2.18 pyodbc 5.3.0 pypandoc 1.16.2 pypdf 6.4.0 PyPDF2 3.0.1 python-calamine 0.6.1 python-docx 1.2.0 python-pptx 1.0.2 pywencai 0.13.1 qianfan 0.4.6 quart-auth 0.11.0 quart-cors 0.8.0 ranx 0.3.20 readability-lxml 0.8.4.1 replicate 0.31.0 reportlab 4.4.6 roman-numbers 1.0.2 ruamel.base 1.0.0 ruamel.yaml 0.18.16 scholarly 1.7.11 selenium-wire 5.1.0 slack_sdk 3.37.0 socksio 1.0.0 sqlglotrs 0.9.0 StrEnum 0.4.15 tavily-python 0.5.1 tencentcloud-sdk-python 3.0.1478 tika 2.6.0 valkey 6.0.2 vertexai 1.70.0 volcengine 1.0.194 voyageai 0.2.3 webdav4 0.10.0 webdriver-manager 4.0.1 wikipedia 1.4.0 word2number 1.1 xgboost 1.6.0 xpinyin 0.7.6 yfinance 0.2.65 zhipuai 2.0.1 ``` ### Type of change - [x] Refactoring Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-16 12:40:03 +08:00
{ name = "socksio" },
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
{ name = "spacy" },
Remove unused py module dependencies (#11964) ### What problem does this PR solve? ``` pip list --not-required Package Version ---------------------------- --------------------- aiosmtplib 5.0.0 akshare 1.17.94 anthropic 0.34.1 arxiv 2.1.3 Aspose.Slides 24.7.0 atlassian-python-api 4.0.7 azure-identity 1.17.1 azure-storage-file-datalake 12.16.0 bio 1.7.1 boxsdk 10.2.0 captcha 0.7.1 cn2an 0.5.22 cohere 5.6.2 Crawl4AI 0.4.247 dashscope 1.20.11 deepl 1.18.0 demjson3 3.0.6 discord.py 2.3.2 dropbox 12.0.2 duckduckgo_search 7.5.5 editdistance 0.8.1 elasticsearch-dsl 8.12.0 exceptiongroup 1.3.1 extract-msg 0.55.0 ffmpeg-python 0.2.0 flasgger 0.9.7.1 Flask-Cors 5.0.0 Flask-Login 0.6.3 Flask-Mail 0.10.0 Flask-Session 0.8.0 google-auth-oauthlib 1.2.3 google-genai 1.55.0 google-generativeai 0.8.5 google_search_results 2.4.2 graspologic 0.1.dev847+g38e680cab groq 0.9.0 grpcio-status 1.67.1 html_text 0.6.2 imageio-ffmpeg 0.6.0 infinity_emb 0.0.66 infinity-sdk 0.6.11 jira 3.10.5 json_repair 0.35.0 langfuse 3.10.5 mammoth 1.11.0 Markdown 3.6 markdown_to_json 2.1.1 markdownify 1.2.2 mcp 1.19.0 mini-racer 0.12.4 minio 7.2.4 mistralai 0.4.2 moodlepy 0.24.1 mypy-boto3-s3 1.40.26 Office365-REST-Python-Client 2.6.2 ollama 0.6.1 onnxruntime-gpu 1.23.2 opencv-python 4.10.0.84 opencv-python-headless 4.10.0.84 opendal 0.45.20 opensearch-py 2.7.1 ormsgpack 1.5.0 pdfplumber 0.10.4 pip 25.3 pluginlib 0.9.4 psycopg2-binary 2.9.11 pyclipper 1.4.0 pycryptodomex 3.20.0 pyobvector 0.2.18 pyodbc 5.3.0 pypandoc 1.16.2 pypdf 6.4.0 PyPDF2 3.0.1 python-calamine 0.6.1 python-docx 1.2.0 python-pptx 1.0.2 pywencai 0.13.1 qianfan 0.4.6 quart-auth 0.11.0 quart-cors 0.8.0 ranx 0.3.20 readability-lxml 0.8.4.1 replicate 0.31.0 reportlab 4.4.6 roman-numbers 1.0.2 ruamel.base 1.0.0 ruamel.yaml 0.18.16 scholarly 1.7.11 selenium-wire 5.1.0 slack_sdk 3.37.0 socksio 1.0.0 sqlglotrs 0.9.0 StrEnum 0.4.15 tavily-python 0.5.1 tencentcloud-sdk-python 3.0.1478 tika 2.6.0 valkey 6.0.2 vertexai 1.70.0 volcengine 1.0.194 voyageai 0.2.3 webdav4 0.10.0 webdriver-manager 4.0.1 wikipedia 1.4.0 word2number 1.1 xgboost 1.6.0 xpinyin 0.7.6 yfinance 0.2.65 zhipuai 2.0.1 ``` ### Type of change - [x] Refactoring Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-16 12:40:03 +08:00
{ name = "sqlglotrs" },
Add tavily as web searh tool. (#5349) ### What problem does this PR solve? #5198 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-02-26 10:21:04 +08:00
{ name = "tavily-python" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "tencentcloud-sdk-python" },
{ name = "tika" },
{ name = "valkey" },
{ name = "volcengine" },
{ name = "voyageai" },
Feat: Add Webdav storage as data source (#11422) ### What problem does this PR solve? This PR adds webdav storage as data source for data sync service. ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-26 07:14:42 +01:00
{ name = "webdav4" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "webdriver-manager" },
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
{ name = "wechatpy" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "wikipedia" },
{ name = "word2number" },
{ name = "xgboost" },
{ name = "xpinyin" },
{ name = "yfinance" },
{ name = "zhipuai" },
]
Feat: sync test group to top pyproject.toml (#8015) ### What problem does this PR solve? sync test group from sdk/python/pyproject.toml to top pyproject.toml ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-06-03 15:21:06 +08:00
[package.dev-dependencies]
test = [
Test: Add code coverage reporting to CI (#12874) ### What problem does this PR solve? Add code coverage reporting to CI ### Type of change - [x] Test (please describe): coverage report --------- Co-authored-by: Liu An <asiro@qq.com>
2026-01-30 14:49:16 +08:00
{ name = "codecov" },
Feat: sync test group to top pyproject.toml (#8015) ### What problem does this PR solve? sync test group from sdk/python/pyproject.toml to top pyproject.toml ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-06-03 15:21:06 +08:00
{ name = "hypothesis" },
{ name = "openpyxl" },
{ name = "pillow" },
Feat: Benchmark CLI additions and documentation (#12536) ### What problem does this PR solve? This PR adds a dedicated HTTP benchmark CLI for RAGFlow chat and retrieval endpoints so we can measure latency/QPS. ### Type of change - [x] Documentation Update - [x] Other (please describe): Adds a CLI benchmarking tool for chat/retrieval latency/QPS --------- Co-authored-by: Liu An <asiro@qq.com>
2026-01-14 13:49:16 +08:00
{ name = "pycryptodomex" },
Feat: sync test group to top pyproject.toml (#8015) ### What problem does this PR solve? sync test group from sdk/python/pyproject.toml to top pyproject.toml ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-06-03 15:21:06 +08:00
{ name = "pytest" },
Add ut test into CI (#11753) ### What problem does this PR solve? As title ### Type of change - [x] Other (please describe): --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-05 11:40:16 +08:00
{ name = "pytest-asyncio" },
{ name = "pytest-cov" },
Update graspologic to gitee (#13362) ### What problem does this PR solve? Accelerate python module downloading ### Type of change - [x] Refactoring Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2026-03-04 17:48:47 +08:00
{ name = "pytest-playwright" },
Add ut test into CI (#11753) ### What problem does this PR solve? As title ### Type of change - [x] Other (please describe): --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-05 11:40:16 +08:00
{ name = "pytest-xdist" },
Feat: sync test group to top pyproject.toml (#8015) ### What problem does this PR solve? sync test group from sdk/python/pyproject.toml to top pyproject.toml ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-06-03 15:21:06 +08:00
{ name = "python-docx" },
{ name = "python-pptx" },
{ name = "reportlab" },
{ name = "requests" },
{ name = "requests-toolbelt" },
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[package.metadata]
requires-dist = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ name = "agentrun-sdk", specifier = ">=0.0.51,<1.0.0" },
build(deps): bump aiohttp from 3.13.3 to 3.14.1 (#16090)
2026-06-16 20:07:32 +08:00
{ name = "aiohttp", specifier = ">=3.14.1" },
Feat: Add box connector (#11845) ### What problem does this PR solve? Feat: Add box connector ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-12 10:23:40 +08:00
{ name = "aiosmtplib", specifier = ">=5.0.0" },
Bump akshare (#4536) ### What problem does this PR solve? Bump akshare. Close #4525 ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-01-20 11:17:59 +08:00
{ name = "akshare", specifier = ">=1.15.78,<2.0.0" },
Feat: add DingTalk AI Table connector and integration for data synch… (#13413) ### What problem does this PR solve? Add DingTalk AI Table connector and integration for data synchronization Issue #13400 ### Type of change - [x] New Feature (non-breaking change which adds functionality) Co-authored-by: wangheyang <wangheyang@corp.netease.com>
2026-03-06 21:13:23 +08:00
{ name = "alibabacloud-dingtalk", specifier = ">=2.0.0" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
{ name = "anthropic", specifier = "==0.76.0" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "arxiv", specifier = "==2.1.3" },
Feat: Add Asana data source integration and configuration options (#12239) ### What problem does this PR solve? change: Add Asana data source integration and configuration options ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2025-12-29 13:28:37 +08:00
{ name = "asana", specifier = ">=5.2.2" },
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
{ name = "atlassian-python-api", specifier = "==4.0.7" },
feat: bump Python minimum from 3.12 to 3.13, drop strenum backport (#14767) Closes #14753 ## What changed | File | Change | |---|---| | `pyproject.toml` | `requires-python` → `>=3.13,<3.15`; remove `strenum==0.4.15` | | `Dockerfile` | `uv python install 3.13`, `uv sync --python 3.13` | | `.github/workflows/tests.yml` | `uv sync --python 3.13` on both matrix legs | | `CLAUDE.md` | dev setup command + requirements note updated | | `deepdoc/parser/mineru_parser.py` | `from strenum import StrEnum` → `from enum import StrEnum` | | `agent/tools/code_exec.py` | same | `StrEnum` has been in the stdlib since Python 3.11 — the `strenum` backport package is no longer needed once the floor is 3.13. ## Why uv.lock is not regenerated `uv lock --python 3.13` fails because: 1. The infiniflow/graspologic fork pins `numpy>=1.26.4,<2.0.0` 2. `tensorflow-cpu>=2.20.0` (the first release with cp313 wheels) depends on `ml-dtypes>=0.5.1`, which requires `numpy>=2.1.0` 3. These two constraints are irreconcilable on Python 3.13 The lockfile regeneration requires loosening the `numpy` upper bound in the `infiniflow/graspologic` fork. Once that fork commit is updated and the SHA in `pyproject.toml:49` is bumped, `uv lock --python 3.13` will succeed. ## RFC corrections Two claims in the original RFC (#14753) did not hold up under code review: - **"graspologic hard-blocks 3.13"** — the infiniflow fork at the pinned commit has no `<3.13` Python constraint. The blocker is the transitive `numpy<2.0.0` conflict with tensorflow-cpu's test dependency, not a direct Python version cap. - **"free-threading throughput gains for I/O-bound workload"** — Python 3.13 free-threading requires a special `--disable-gil` build and provides no benefit for async I/O code (the GIL is already released during I/O). The real motivation is forward compatibility and improved error messages.
2026-05-15 08:40:53 +02:00
{ name = "audioop-lts", specifier = ">=0.2.1" },
Bump to infinity v0.7.0-dev6 (#14606) ### What problem does this PR solve? Bump to infinity v0.7.0-dev6 (uv lock --upgrade-package infinity-sdk) ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-07 10:51:17 +08:00
{ name = "azure-identity", specifier = ">=1.25.3" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "azure-storage-file-datalake", specifier = "==12.16.0" },
Bump beartype to 0.22.6 (#11463) ### What problem does this PR solve? Bump beartype to 0.22.6 ### Type of change - [x] Refactoring
2025-11-22 11:56:43 +08:00
{ name = "beartype", specifier = ">=0.20.0,<1.0.0" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "bio", specifier = "==1.7.1" },
Feat: add BedrockCV for vision/image2text inference via LiteLLM (#14705) ## Summary - `CvModel["Bedrock"]` was absent from `rag/llm/cv_model.py`, causing `model_instance()` to return `None` when a Bedrock model was used as a PDF parser — even after correct model resolution. - This PR adds `BedrockCV`, enabling Bedrock vision models (e.g. `amazon.nova-pro-v1:0`, `anthropic.claude-3-5-sonnet`) to be used as PDF parsers. ## What problem does this PR solve? When a Bedrock model is selected as the PDF parser in a knowledge base, ingestion failed with: ``` 'LiteLLMBase' object has no attribute 'describe_with_prompt' ``` The root cause: `LiteLLMBase` (the Bedrock chat implementation) was the only registered handler for the Bedrock factory. It does not implement `describe_with_prompt`. `CvModel` had no Bedrock entry, so `model_instance()` returned `None` for `image2text` requests. ## Type of change - [x] New Feature (non-breaking change which adds functionality) ## Changes **`rag/llm/cv_model.py`** Adds `BedrockCV(Base)` with `_FACTORY_NAME = "Bedrock"`: - Uses `litellm.completion` with the `bedrock/` prefix (consistent with `LiteLLMBase`) - Parses AWS credentials from the JSON key assembled by `add_llm` (`auth_mode`, `bedrock_ak`, `bedrock_sk`, `bedrock_region`, `aws_role_arn`) - Supports three auth modes: `access_key_secret`, `iam_role` (via STS `assume_role`), and default credential chain (IRSA, instance profile) - Implements `describe_with_prompt` and `describe` ## Test plan - [ ] Configure a Bedrock vision model (e.g. `amazon.nova-pro-v1:0`) with valid AWS credentials - [ ] Select it as PDF parser in a knowledge base - [ ] Verify ingestion of a PDF document completes without errors - [ ] Verify `CvModel["Bedrock"]` resolves to `BedrockCV` 🤖 Generated with [Claude Code](https://claude.ai/claude-code) --------- Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-05-11 04:29:58 +02:00
{ name = "boto3", specifier = ">=1.28.0" },
Feat: Add box connector (#11845) ### What problem does this PR solve? Feat: Add box connector ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-12 10:23:40 +08:00
{ name = "boxsdk", specifier = ">=10.1.0" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
{ name = "browser-use", specifier = ">=0.11.1,<0.12.0" },
Feat: add forgot password reset, solve #8547 (#10586) ### What problem does this PR solve? Feat: add forgot password reset, solve #8547 ### Type of change - [X] New Feature (non-breaking change which adds functionality)
2025-10-16 15:07:49 +08:00
{ name = "captcha", specifier = ">=0.7.1" },
Feat: optimize ingestion pipeline with preprocess (#13211) ### What problem does this PR solve? Feat: optimize ingestion pipeline with preprocess ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-02-26 10:24:13 +08:00
{ name = "chardet", specifier = ">=5.2.0,<6.0.0" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "cn2an", specifier = "==0.5.22" },
{ name = "cohere", specifier = "==5.6.2" },
build(deps): bump crawl4ai from 0.8.9 to 0.9.0 (#16470)
2026-06-30 09:34:48 +08:00
{ name = "crawl4ai", specifier = ">=0.9.0,<0.9.1" },
Chore/upgrade dashscope to 1.25.11 (#13007) ## Description Upgrade dashscope package to support text-embedding-v4 model. ## Changes - Update dashscope version from 1.20.11 to 1.25.11 in pyproject.toml ## Reason The text-embedding-v4 model requires dashscope >= 1.25.0 to function properly. This upgrade ensures compatibility with the latest embedding models. Co-authored-by: Clint-chan <Clint-chan@users.noreply.github.com>
2026-02-06 19:06:41 +08:00
{ name = "dashscope", specifier = "==1.25.11" },
Go: add balance command (#14262) ### What problem does this PR solve? ``` RAGFlow(user)> list supported models from 'moonshot' 'test'; +---------------------------------+ | model_name | +---------------------------------+ | moonshot-v1-32k-vision-preview | | kimi-k2.6 | | moonshot-v1-8k | | moonshot-v1-auto | | moonshot-v1-128k | | moonshot-v1-32k | | kimi-k2.5 | | moonshot-v1-8k-vision-preview | | moonshot-v1-128k-vision-preview | +---------------------------------+ RAGFlow(user)> show balance from 'moonshot' 'test'; +---------+----------+ | balance | currency | +---------+----------+ | 0 | CNY | +---------+----------+ ``` ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2026-04-21 21:31:50 +08:00
{ name = "debugpy", specifier = ">=1.8.13" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "deepl", specifier = "==1.18.0" },
{ name = "demjson3", specifier = "==3.0.6" },
{ name = "discord-py", specifier = "==2.3.2" },
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
{ name = "discord-py", specifier = ">=2.3.0" },
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
{ name = "dropbox", specifier = "==12.0.2" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "duckduckgo-search", specifier = ">=7.2.0,<8.0.0" },
{ name = "editdistance", specifier = "==0.8.1" },
{ name = "elasticsearch-dsl", specifier = "==8.12.0" },
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
{ name = "en-core-web-sm", url = "https://github.com/explosion/spacy-models/releases/download/en_core_web_sm-3.8.0/en_core_web_sm-3.8.0-py3-none-any.whl" },
Use Infinity single-field-multi-index (#11444) ### What problem does this PR solve? Use Infinity single-field-multi-index ### Type of change - [x] Refactoring - [x] Performance Improvement
2025-11-26 11:06:37 +08:00
{ name = "exceptiongroup", specifier = ">=1.3.0,<2.0.0" },
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
{ name = "extract-msg", specifier = ">=0.39.0" },
feat: support rss datasource (#13721) ### What problem does this PR solve? Supporting public RSS/Atom feed URLs as data sources for RagFlow. link https://github.com/infiniflow/ragflow/issues/12313 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-03-27 22:58:44 +08:00
{ name = "feedparser", specifier = ">=6.0.11,<7.0.0" },
Feat:new api /sequence2txt and update QWenSeq2txt (#11643) ### What problem does this PR solve? change: new api /sequence2txt, update QWenSeq2txt and ZhipuSeq2txt ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-02 11:17:31 +08:00
{ name = "ffmpeg-python", specifier = ">=0.2.0" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "flasgger", specifier = ">=0.9.7.1,<0.10.0" },
Fix security issue (#11965) ### What problem does this PR solve? - CVE-2024-6866 - CVE-2024-6844 - CVE-2024-6839 ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue) Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-16 13:31:45 +08:00
{ name = "flask-cors", specifier = "==6.0.2" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "flask-login", specifier = "==0.6.3" },
Feat: add SMTP support for user invitation emails (#9479) ### What problem does this PR solve? Add SMTP support for user invitation emails ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-08-15 18:12:20 +08:00
{ name = "flask-mail", specifier = ">=0.10.0" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "flask-session", specifier = "==0.8.0" },
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
{ name = "google-api-python-client", specifier = ">=2.190.0,<3.0.0" },
Feat: add initial Google Drive connector support (#11147) ### What problem does this PR solve? This feature is primarily ported from the [Onyx](https://github.com/onyx-dot-app/onyx) project with necessary modifications. Thanks for such a brilliant project. Minor: consistently use `google_drive` rather than `google_driver`. <img width="566" height="731" alt="image" src="https://github.com/user-attachments/assets/6f64e70e-881e-42c7-b45f-809d3e0024a4" /> <img width="904" height="830" alt="image" src="https://github.com/user-attachments/assets/dfa7d1ef-819a-4a82-8c52-0999f48ed4a6" /> <img width="911" height="869" alt="image" src="https://github.com/user-attachments/assets/39e792fb-9fbe-4f3d-9b3c-b2265186bc22" /> <img width="947" height="323" alt="image" src="https://github.com/user-attachments/assets/27d70e96-d9c0-42d9-8c89-276919b6d61d" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-10 19:15:02 +08:00
{ name = "google-auth-oauthlib", specifier = ">=1.2.0,<2.0.0" },
build(deps): bump crawl4ai from 0.8.9 to 0.9.0 (#16470)
2026-06-30 09:34:48 +08:00
{ name = "google-cloud-bigquery", specifier = ">=3.25.0,<4.0.0" },
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
{ name = "google-cloud-storage", specifier = ">=2.19.0,<3.0.0" },
Fix problem with Google Cloud models with reasoning (like gemini) - Additional fix to issue #10474 (#10502) ### What problem does this PR solve? Issue #10474 - Update to PR #10477 ### Type of change - [X] Bug Fix (non-breaking change which fixes an issue)
2025-10-15 08:54:20 +02:00
{ name = "google-genai", specifier = ">=1.41.0,<2.0.0" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "google-search-results", specifier = "==2.4.2" },
Go: refactor embedding interface (#14757) ### What problem does this PR solve? Provide embedding index according to the input text ### Type of change - [x] Refactoring --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2026-05-11 14:45:30 +08:00
{ name = "graspologic", git = "https://gitee.com/infiniflow/graspologic.git?rev=38e680cab72bc9fb68a7992c3bcc2d53b24e42fd" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
{ name = "groq", specifier = ">=0.30.0,<1.0.0" },
Remove unused py module dependencies (#11964) ### What problem does this PR solve? ``` pip list --not-required Package Version ---------------------------- --------------------- aiosmtplib 5.0.0 akshare 1.17.94 anthropic 0.34.1 arxiv 2.1.3 Aspose.Slides 24.7.0 atlassian-python-api 4.0.7 azure-identity 1.17.1 azure-storage-file-datalake 12.16.0 bio 1.7.1 boxsdk 10.2.0 captcha 0.7.1 cn2an 0.5.22 cohere 5.6.2 Crawl4AI 0.4.247 dashscope 1.20.11 deepl 1.18.0 demjson3 3.0.6 discord.py 2.3.2 dropbox 12.0.2 duckduckgo_search 7.5.5 editdistance 0.8.1 elasticsearch-dsl 8.12.0 exceptiongroup 1.3.1 extract-msg 0.55.0 ffmpeg-python 0.2.0 flasgger 0.9.7.1 Flask-Cors 5.0.0 Flask-Login 0.6.3 Flask-Mail 0.10.0 Flask-Session 0.8.0 google-auth-oauthlib 1.2.3 google-genai 1.55.0 google-generativeai 0.8.5 google_search_results 2.4.2 graspologic 0.1.dev847+g38e680cab groq 0.9.0 grpcio-status 1.67.1 html_text 0.6.2 imageio-ffmpeg 0.6.0 infinity_emb 0.0.66 infinity-sdk 0.6.11 jira 3.10.5 json_repair 0.35.0 langfuse 3.10.5 mammoth 1.11.0 Markdown 3.6 markdown_to_json 2.1.1 markdownify 1.2.2 mcp 1.19.0 mini-racer 0.12.4 minio 7.2.4 mistralai 0.4.2 moodlepy 0.24.1 mypy-boto3-s3 1.40.26 Office365-REST-Python-Client 2.6.2 ollama 0.6.1 onnxruntime-gpu 1.23.2 opencv-python 4.10.0.84 opencv-python-headless 4.10.0.84 opendal 0.45.20 opensearch-py 2.7.1 ormsgpack 1.5.0 pdfplumber 0.10.4 pip 25.3 pluginlib 0.9.4 psycopg2-binary 2.9.11 pyclipper 1.4.0 pycryptodomex 3.20.0 pyobvector 0.2.18 pyodbc 5.3.0 pypandoc 1.16.2 pypdf 6.4.0 PyPDF2 3.0.1 python-calamine 0.6.1 python-docx 1.2.0 python-pptx 1.0.2 pywencai 0.13.1 qianfan 0.4.6 quart-auth 0.11.0 quart-cors 0.8.0 ranx 0.3.20 readability-lxml 0.8.4.1 replicate 0.31.0 reportlab 4.4.6 roman-numbers 1.0.2 ruamel.base 1.0.0 ruamel.yaml 0.18.16 scholarly 1.7.11 selenium-wire 5.1.0 slack_sdk 3.37.0 socksio 1.0.0 sqlglotrs 0.9.0 StrEnum 0.4.15 tavily-python 0.5.1 tencentcloud-sdk-python 3.0.1478 tika 2.6.0 valkey 6.0.2 vertexai 1.70.0 volcengine 1.0.194 voyageai 0.2.3 webdav4 0.10.0 webdriver-manager 4.0.1 wikipedia 1.4.0 word2number 1.1 xgboost 1.6.0 xpinyin 0.7.6 yfinance 0.2.65 zhipuai 2.0.1 ``` ### Type of change - [x] Refactoring Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-16 12:40:03 +08:00
{ name = "grpcio-status", specifier = "==1.67.1" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "html-text", specifier = "==0.6.2" },
{ name = "infinity-emb", specifier = ">=0.0.66,<0.0.67" },
Bump to infinity v0.7.0 (#14968) ### What problem does this PR solve? Upgrade infinity ### Type of change - [x] Refactoring
2026-05-18 10:25:59 +08:00
{ name = "infinity-sdk", specifier = "==0.7.0" },
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
{ name = "jira", specifier = "==3.10.5" },
Light GraphRAG (#4585) ### What problem does this PR solve? #4543 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-01-22 19:43:14 +08:00
{ name = "json-repair", specifier = "==0.35.0" },
fix: migrate Langfuse integration from start_generation to start_obse… (#14205) The Langfuse Python SDK v3+ removed `start_generation()` method. RagFlow's code called this non-existent method, causing AttributeError when Langfuse tracing is enabled. Replace all `start_generation()` calls with `start_observation(as_type="generation")` which is the correct v4 SDK API. Affected files: - api/db/services/llm_service.py (12 occurrences) - api/db/services/dialog_service.py (1 occurrence) Fixes #14204 Related to #9243 ### What problem does this PR solve? _Briefly describe what this PR aims to solve. Include background context that will help reviewers understand the purpose of the PR._ ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue) --------- Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
2026-04-24 06:03:57 +04:00
{ name = "langfuse", specifier = ">=4.0.1" },
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
{ name = "lark-oapi", specifier = ">=1.2.0" },
{ name = "line-bot-sdk", specifier = ">=3.0.0" },
feat(agent): Go port — canvas engine, 22 components, DSL v2, 13 endpoints (#15952) Ports the agent canvas subsystem from Python to Go. ## What's included ### Canvas Engine (Phase 0/1) - State engine, scheduler, variable resolver, Redis checkpoint store, cancel protocol - **209 tests** across canvas / component / io packages ### 22 Components (P0–P4) | Tier | Components | |---|---| | P0 T1+T2+T3 | LLM, Agent, ExitLoop, Switch, Categorize, Begin, Message, Invoke | | P1 T3 | VariableAggregator, VariableAssigner, StringTransform, ListOperations, DataOperations | | P2 T3 | Iteration, IterationItem, Loop, LoopItem | | P3 T3 | UserFillUp, Fillup | | P4 T5 | Browser, ExcelProcessor, DocsGenerator | ### DSL v2 Schema (Phase 2.5) - Typed v2 in-memory model with v1-to-v2 auto-detect converter - v1 legacy field stripping per plan §2.11.7 ### HTTP Endpoints & Bug Fixes (Plans PR1–PR3) - **DELETE SQL bug fix**: gorm v2 `Where("id = ?", id).Delete(...)` pattern - **CreateAgent validation**: title/DSL required, duplicate check, 103 envelope - **13 new endpoints**: templates, prompts, tags, sessions CRUD, chat/completions (SSE + non-stream stubs), rerun, test_db_connection, logs, webhook/logs - **756 Go unit tests** (745 → 756, +18) - **17 → 0 Python integration test failures** (test_agents.py + test_session_management/) ### Tools 21 eino tools: HTTPHelper, search tools, financial/data tools, mandatory stubs ### Infrastructure OTel observability, NATS message queue, DeepDoc gRPC client, SSRF guards, IDOR mitigation
2026-06-12 22:58:28 +08:00
{ name = "litellm", specifier = "==1.82.5" },
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
{ name = "mammoth", specifier = ">=1.11.0" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "markdown", specifier = "==3.6" },
{ name = "markdown-to-json", specifier = "==2.1.1" },
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
{ name = "markdownify", specifier = ">=1.2.0" },
Fix security issue (#11965) ### What problem does this PR solve? - CVE-2024-6866 - CVE-2024-6844 - CVE-2024-6839 ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue) Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-16 13:31:45 +08:00
{ name = "mcp", specifier = ">=1.19.0" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "mini-racer", specifier = ">=0.12.4,<0.13.0" },
{ name = "minio", specifier = "==7.2.4" },
{ name = "mistralai", specifier = "==0.4.2" },
Add Moodle data source integration (#11325) ### What problem does this PR solve? This PR adds a native Moodle connector to sync content (courses, resources, forums, assignments, pages, books) into RAGFlow. ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2025-11-21 12:58:49 +01:00
{ name = "moodlepy", specifier = ">=0.23.0" },
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
{ name = "mypy-boto3-s3", specifier = "==1.40.26" },
feat/add MySQL and PostgreSQL data source connectors (#12817) ### What problem does this PR solve? This PR adds MySQL and PostgreSQL as data source connectors, allowing users to import data directly from relational databases into RAGFlow for RAG workflows. Many users store their knowledge in databases (product catalogs, documentation, FAQs, etc.) and currently have no way to sync this data into RAGFlow without exporting to files first. This feature lets them connect directly to their databases, run SQL queries, and automatically create documents from the results. Closes #763 Closes #11560 ### Type of change - [ ] Bug Fix (non-breaking change which fixes an issue) - [x] New Feature (non-breaking change which adds functionality) - [ ] Documentation Update - [ ] Refactoring - [ ] Performance Improvement - [ ] Other (please describe): ### What this PR does **New capabilities:** - Connect to MySQL and PostgreSQL databases - Run custom SQL queries to extract data - Map database columns to document content (vectorized) and metadata (searchable) - Support incremental sync using a timestamp column - Full frontend UI with connection form and tooltips **Files changed:** Backend: - `common/constants.py` - Added MYSQL/POSTGRESQL to FileSource enum - `common/data_source/config.py` - Added to DocumentSource enum - `common/data_source/rdbms_connector.py` - New connector (368 lines) - `common/data_source/__init__.py` - Exported the connector - `rag/svr/sync_data_source.py` - Added MySQL and PostgreSQL sync classes - `pyproject.toml` - Added mysql-connector-python dependency Frontend: - `web/src/pages/user-setting/data-source/constant/index.tsx` - Form fields - `web/src/locales/en.ts` - English translations - `web/src/assets/svg/data-source/mysql.svg` - MySQL icon - `web/src/assets/svg/data-source/postgresql.svg` - PostgreSQL icon ### Testing done Tested with MySQL 8.0 and PostgreSQL 16: - Connection validation works correctly - Full sync imports all query results as documents - Incremental sync only fetches rows updated since last sync - Custom SQL queries filter data as expected - Invalid credentials show clear error messages - Lint checks pass (`ruff check` returns no errors) --------- Co-authored-by: mkdev11 <YOUR_GITHUB_ID+MkDev11@users.noreply.github.com>
2026-02-03 23:14:32 -03:00
{ name = "mysql-connector-python", specifier = ">=9.0.0,<10.0.0" },
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
{ name = "nest-asyncio", specifier = ">=1.6.0,<2.0.0" },
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
{ name = "office365-rest-python-client", specifier = "==2.6.2" },
Fix problem with Google Cloud models with reasoning (like gemini) - Additional fix to issue #10474 (#10502) ### What problem does this PR solve? Issue #10474 - Update to PR #10477 ### Type of change - [X] Bug Fix (non-breaking change which fixes an issue)
2025-10-15 08:54:20 +02:00
{ name = "ollama", specifier = ">=0.5.0" },
Bump python to >=3.12 (#11846) ### What problem does this PR solve? Bump python to >=3.12 ### Type of change - [x] Refactoring
2025-12-09 19:55:25 +08:00
{ name = "onnxruntime", marker = "platform_machine != 'x86_64' or sys_platform == 'darwin'", specifier = "==1.23.2" },
{ name = "onnxruntime-gpu", marker = "platform_machine == 'x86_64' and sys_platform != 'darwin'", specifier = "==1.23.2" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "opencv-python", specifier = "==4.10.0.84" },
{ name = "opencv-python-headless", specifier = "==4.10.0.84" },
Oss support opendal(including mysql) (#8204) ### What problem does this PR solve? #8074 Oss support opendal(including mysql) ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2025-06-12 11:37:42 +08:00
{ name = "opendal", specifier = ">=0.45.0,<0.46.0" },
Feat: Adds OpenSearch2.19.1 as the vector_database support (#7140) ### What problem does this PR solve? This PR adds the support for latest OpenSearch2.19.1 as the store engine & search engine option for RAGFlow. ### Main Benefit 1. OpenSearch2.19.1 is licensed under the [Apache v2.0 License] which is much better than Elasticsearch 2. For search, OpenSearch2.19.1 supports full-text search、vector_search、hybrid_search those are similar with Elasticsearch on schema 3. For store, OpenSearch2.19.1 stores text、vector those are quite simliar with Elasticsearch on schema ### Changes - Support opensearch_python_connetor. I make a lot of adaptions since the schema and api/method between ES and Opensearch differs in many ways(especially the knn_search has a significant gap) : rag/utils/opensearch_coon.py - Support static config adaptions by changing: conf/service_conf.yaml、api/settings.py、rag/settings.py - Supprt some store&search schema changes between OpenSearch and ES: conf/os_mapping.json - Support OpenSearch python sdk : pyproject.toml - Support docker config for OpenSearch2.19.1 : docker/.env、docker/docker-compose-base.yml、docker/service_conf.yaml.template ### How to use - I didn't change the priority that ES as the default doc/search engine. Only if in docker/.env , we set DOC_ENGINE=${DOC_ENGINE:-opensearch}, it will work. ### Others Our team tested a lot of docs in our environment by using OpenSearch as the vector database ,it works very well. All the conifg for OpenSearch is necessary. ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Yongteng Lei <yongtengrey@outlook.com> Co-authored-by: writinwaters <93570324+writinwaters@users.noreply.github.com> Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
2025-04-24 16:03:31 +08:00
{ name = "opensearch-py", specifier = "==2.7.1" },
fix: upgrade crawl4ai to 0.8.0 (CVE-2026-26217) (#15415) ## Summary Upgrade crawl4ai from 0.7.6 to 0.8.0 to fix CVE-2026-26217. ## Vulnerability | Field | Value | |-------|-------| | **ID** | CVE-2026-26217 | | **Severity** | CRITICAL | | **Scanner** | trivy | | **Rule** | `CVE-2026-26217` | | **File** | `uv.lock` | | **Assessment** | Likely exploitable | **Description**: Crawl4AI Has Local File Inclusion in Docker API via file:// URLs ## Evidence **Scanner confirmation**: trivy rule `CVE-2026-26217` flagged this pattern. **Production code**: This file is in the production codebase, not test-only code. ## Threat Model Context This is a web service - vulnerabilities in request handlers are directly exploitable by remote attackers. ## Changes - `pyproject.toml` - `uv.lock` ## Verification - [x] Build passes - [x] Scanner re-scan confirms fix - [x] LLM code review passed --- *This change addresses a pattern flagged by static analysis. The code path handles user-influenced input and the fix reduces the attack surface against both manual and automated exploitation.* --- *Automated security fix by [OrbisAI Security](https://orbisappsec.com)*
2026-05-29 19:08:41 +05:30
{ name = "ormsgpack", specifier = ">=1.6.0" },
Feat: add local & ssh provider in admin panel (#15039) ### What problem does this PR solve? Feat: add local & ssh provider in admin panel ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-20 16:56:20 +08:00
{ name = "paramiko", specifier = ">=3.5.1" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "pdfplumber", specifier = "==0.10.4" },
Feat: optimize ingestion pipeline with preprocess (#13211) ### What problem does this PR solve? Feat: optimize ingestion pipeline with preprocess ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-02-26 10:24:13 +08:00
{ name = "peewee", specifier = ">=3.17.1,<4.0.0" },
Bump to infinity v0.7.0-dev6 (#14606) ### What problem does this PR solve? Bump to infinity v0.7.0-dev6 (uv lock --upgrade-package infinity-sdk) ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-07 10:51:17 +08:00
{ name = "pluginlib", specifier = ">=0.10.0" },
Bump python to >=3.12 (#11846) ### What problem does this PR solve? Bump python to >=3.12 ### Type of change - [x] Refactoring
2025-12-09 19:55:25 +08:00
{ name = "psycopg2-binary", specifier = ">=2.9.11,<3.0.0" },
Feat: add Airtable connector and integration for data synchronization (#12211) ### What problem does this PR solve? change: add Airtable connector and integration for data synchronization ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-25 17:50:41 +08:00
{ name = "pyairtable", specifier = ">=3.3.0" },
Bump python to >=3.12 (#11846) ### What problem does this PR solve? Bump python to >=3.12 ### Type of change - [x] Refactoring
2025-12-09 19:55:25 +08:00
{ name = "pyclipper", specifier = ">=1.4.0,<2.0.0" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "pycryptodomex", specifier = "==3.20.0" },
Feat: GitHub connector (#12314) ### What problem does this PR solve? Feat: GitHub connector ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-12-30 15:09:52 +08:00
{ name = "pygithub", specifier = ">=2.8.1" },
chore: bump pyobvector from 0.2.18 to 0.2.22 (#12640) ### What problem does this PR solve? Update ob client ### Type of change - [x] Other (please describe):dependency upgrade
2026-01-15 15:21:34 +08:00
{ name = "pyobvector", specifier = "==0.2.22" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "pyodbc", specifier = ">=5.2.0,<6.0.0" },
Feat: extract message output to file (#11251) ### What problem does this PR solve? Feat: extract message output to file ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-14 19:52:11 +08:00
{ name = "pypandoc", specifier = ">=1.16" },
Build(deps): Bump pypdf from 6.9.2 to 6.10.2 (#14184) Bumps [pypdf](https://github.com/py-pdf/pypdf) from 6.9.2 to 6.10.2. <details> <summary>Release notes</summary> <p><em>Sourced from <a href="https://github.com/py-pdf/pypdf/releases">pypdf's releases</a>.</em></p> <blockquote> <h2>Version 6.10.2, 2026-04-15</h2> <h2>What's new</h2> <h3>Security (SEC)</h3> <ul> <li>Do not rely on possibly invalid /Size for incremental cloning (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3735">#3735</a>) by <a href="https://github.com/stefan6419846"><code>@​stefan6419846</code></a></li> <li>Introduce limits for FlateDecode parameters and image decoding (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3734">#3734</a>) by <a href="https://github.com/stefan6419846"><code>@​stefan6419846</code></a></li> </ul> <p><a href="https://github.com/py-pdf/pypdf/compare/6.10.1...6.10.2">Full Changelog</a></p> <h2>Version 6.10.1, 2026-04-14</h2> <h2>What's new</h2> <h3>Security (SEC)</h3> <ul> <li>Limit the allowed size of xref and object streams (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3733">#3733</a>) by <a href="https://github.com/stefan6419846"><code>@​stefan6419846</code></a></li> </ul> <h3>Robustness (ROB)</h3> <ul> <li>Consider strict mode setting for decryption errors (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3731">#3731</a>) by <a href="https://github.com/stefan6419846"><code>@​stefan6419846</code></a></li> </ul> <h3>Documentation (DOC)</h3> <ul> <li>Use new parameter names for compress_identical_objects by <a href="https://github.com/stefan6419846"><code>@​stefan6419846</code></a></li> </ul> <p><a href="https://github.com/py-pdf/pypdf/compare/6.10.0...6.10.1">Full Changelog</a></p> <h2>Version 6.10.0, 2026-04-10</h2> <h2>What's new</h2> <h3>Security (SEC)</h3> <ul> <li>Disallow custom XML entity declarations for XMP metadata (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3724">#3724</a>) by <a href="https://github.com/stefan6419846"><code>@​stefan6419846</code></a></li> </ul> <h3>New Features (ENH)</h3> <ul> <li>Skip MD5 key derivation for AES-256 encrypted PDFs (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3694">#3694</a>) by <a href="https://github.com/Ygnas"><code>@​Ygnas</code></a></li> </ul> <h3>Bug Fixes (BUG)</h3> <ul> <li>Use remove_orphans in compress_identical_objects (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3310">#3310</a>) by <a href="https://github.com/j-t-1"><code>@​j-t-1</code></a></li> <li>Fix PdfReadError when xref table contains comments before trailer (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3710">#3710</a>) by <a href="https://github.com/rassie"><code>@​rassie</code></a></li> <li>Correctly verify AES padding during decryption (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3699">#3699</a>) by <a href="https://github.com/stefan6419846"><code>@​stefan6419846</code></a></li> <li>Fix stale object cache from non-authoritative object streams (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3698">#3698</a>) by <a href="https://github.com/astahlman"><code>@​astahlman</code></a></li> <li>Fix extract_links pairing when annotations include non-links (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3687">#3687</a>) by <a href="https://github.com/ReinerBRO"><code>@​ReinerBRO</code></a></li> </ul> <h3>Documentation (DOC)</h3> <ul> <li>Add AI policy (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3717">#3717</a>) by <a href="https://github.com/stefan6419846"><code>@​stefan6419846</code></a></li> </ul> <p><a href="https://github.com/py-pdf/pypdf/compare/6.9.2...6.10.0">Full Changelog</a></p> </blockquote> </details> <details> <summary>Changelog</summary> <p><em>Sourced from <a href="https://github.com/py-pdf/pypdf/blob/main/CHANGELOG.md">pypdf's changelog</a>.</em></p> <blockquote> <h2>Version 6.10.2, 2026-04-15</h2> <h3>Security (SEC)</h3> <ul> <li>Do not rely on possibly invalid /Size for incremental cloning (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3735">#3735</a>)</li> <li>Introduce limits for FlateDecode parameters and image decoding (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3734">#3734</a>)</li> </ul> <p><a href="https://github.com/py-pdf/pypdf/compare/6.10.1...6.10.2">Full Changelog</a></p> <h2>Version 6.10.1, 2026-04-14</h2> <h3>Security (SEC)</h3> <ul> <li>Limit the allowed size of xref and object streams (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3733">#3733</a>)</li> </ul> <h3>Robustness (ROB)</h3> <ul> <li>Consider strict mode setting for decryption errors (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3731">#3731</a>)</li> </ul> <h3>Documentation (DOC)</h3> <ul> <li>Use new parameter names for compress_identical_objects</li> </ul> <p><a href="https://github.com/py-pdf/pypdf/compare/6.10.0...6.10.1">Full Changelog</a></p> <h2>Version 6.10.0, 2026-04-10</h2> <h3>Security (SEC)</h3> <ul> <li>Disallow custom XML entity declarations for XMP metadata (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3724">#3724</a>)</li> </ul> <h3>New Features (ENH)</h3> <ul> <li>Skip MD5 key derivation for AES-256 encrypted PDFs (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3694">#3694</a>)</li> </ul> <h3>Bug Fixes (BUG)</h3> <ul> <li>Use remove_orphans in compress_identical_objects (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3310">#3310</a>)</li> <li>Fix PdfReadError when xref table contains comments before trailer (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3710">#3710</a>)</li> <li>Correctly verify AES padding during decryption (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3699">#3699</a>)</li> <li>Fix stale object cache from non-authoritative object streams (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3698">#3698</a>)</li> <li>Fix extract_links pairing when annotations include non-links (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3687">#3687</a>)</li> </ul> <h3>Documentation (DOC)</h3> <ul> <li>Add AI policy (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3717">#3717</a>)</li> </ul> <p><a href="https://github.com/py-pdf/pypdf/compare/6.9.2...6.10.0">Full Changelog</a></p> </blockquote> </details> <details> <summary>Commits</summary> <ul> <li><a href="https://github.com/py-pdf/pypdf/commit/c476b4f293c8ef4cac07dfb755e5582d838fcdc0"><code>c476b4f</code></a> REL: 6.10.2</li> <li><a href="https://github.com/py-pdf/pypdf/commit/c50a0104cf083356f7c7f5d61410466a57f5c88a"><code>c50a010</code></a> SEC: Do not rely on possibly invalid /Size for incremental cloning (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3735">#3735</a>)</li> <li><a href="https://github.com/py-pdf/pypdf/commit/ac734dab4eef92bcce50d503949b4d9887d89f11"><code>ac734da</code></a> SEC: Introduce limits for FlateDecode parameters and image decoding (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3734">#3734</a>)</li> <li><a href="https://github.com/py-pdf/pypdf/commit/b49e7eb45422c19b68ac59c51b7699409e74d44e"><code>b49e7eb</code></a> REL: 6.10.1</li> <li><a href="https://github.com/py-pdf/pypdf/commit/62338e9d36419cf193ccec7331784f45df1d70b3"><code>62338e9</code></a> SEC: Limit the allowed size of xref and object streams (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3733">#3733</a>)</li> <li><a href="https://github.com/py-pdf/pypdf/commit/5dcc0aebaa2c732028ea8def2eb9982e324b7c11"><code>5dcc0ae</code></a> DEV: Update pytest-benchmark to 5.2.3</li> <li><a href="https://github.com/py-pdf/pypdf/commit/b42e4aa98ae5c7fdd02558d165d39fe639fdf97d"><code>b42e4aa</code></a> DEV: Update pinned pillow and pytest where possible (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3732">#3732</a>)</li> <li><a href="https://github.com/py-pdf/pypdf/commit/717446b1218a3eb236cb47d1bae2b68451ccb6c0"><code>717446b</code></a> ROB: Consider strict mode setting for decryption errors (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3731">#3731</a>)</li> <li><a href="https://github.com/py-pdf/pypdf/commit/9e461d361b9004da68fc8e6acc4308cce68aa304"><code>9e461d3</code></a> DEV: Bump softprops/action-gh-release from 2 to 3 (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3730">#3730</a>)</li> <li><a href="https://github.com/py-pdf/pypdf/commit/500d09d92fa80a6f1fcdfa46656893efd05e91ff"><code>500d09d</code></a> TST: Update <code>test_embedded_file__basic</code> to use <code>tmp_path</code> fixture (<a href="https://redirect.github.com/py-pdf/pypdf/issues/3726">#3726</a>)</li> <li>Additional commits viewable in <a href="https://github.com/py-pdf/pypdf/compare/6.9.2...6.10.2">compare view</a></li> </ul> </details> <br /> [![Dependabot compatibility score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=pypdf&package-manager=uv&previous-version=6.9.2&new-version=6.10.2)](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores) Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting `@dependabot rebase`. [//]: # (dependabot-automerge-start) [//]: # (dependabot-automerge-end) --- <details> <summary>Dependabot commands and options</summary> <br /> You can trigger Dependabot actions by commenting on this PR: - `@dependabot rebase` will rebase this PR - `@dependabot recreate` will recreate this PR, overwriting any edits that have been made to it - `@dependabot show <dependency name> ignore conditions` will show all of the ignore conditions of the specified dependency - `@dependabot ignore this major version` will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself) - `@dependabot ignore this minor version` will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself) - `@dependabot ignore this dependency` will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself) You can disable automated security fix PRs for this repo from the [Security Alerts page](https://github.com/infiniflow/ragflow/network/alerts). </details> Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2026-04-17 18:43:19 +08:00
{ name = "pypdf", specifier = ">=6.10.2" },
Add fallback to use 'calamine' parse engine in excel_parser.py (#9374) ### What problem does this PR solve? add fallback to `calamine` engine when parse error raised using the default `openpyxl` / `xlrd` engine. e.g. the following error can be fixed: ``` Traceback (most recent call last): File "/ragflow/deepdoc/parser/excel_parser.py", line 53, in _load_excel_to_workbook df = pd.read_excel(file_like_object) File "/ragflow/.venv/lib/python3.10/site-packages/pandas/io/excel/_base.py", line 495, in read_excel io = ExcelFile( File "/ragflow/.venv/lib/python3.10/site-packages/pandas/io/excel/_base.py", line 1567, in __init__ self._reader = self._engines[engine]( File "/ragflow/.venv/lib/python3.10/site-packages/pandas/io/excel/_xlrd.py", line 46, in __init__ super().__init__( File "/ragflow/.venv/lib/python3.10/site-packages/pandas/io/excel/_base.py", line 573, in __init__ self.book = self.load_workbook(self.handles.handle, engine_kwargs) File "/ragflow/.venv/lib/python3.10/site-packages/pandas/io/excel/_xlrd.py", line 63, in load_workbook return open_workbook(file_contents=data, **engine_kwargs) File "/ragflow/.venv/lib/python3.10/site-packages/xlrd/__init__.py", line 172, in open_workbook bk = open_workbook_xls( File "/ragflow/.venv/lib/python3.10/site-packages/xlrd/book.py", line 68, in open_workbook_xls bk.biff2_8_load( File "/ragflow/.venv/lib/python3.10/site-packages/xlrd/book.py", line 641, in biff2_8_load cd.locate_named_stream(UNICODE_LITERAL(qname)) File "/ragflow/.venv/lib/python3.10/site-packages/xlrd/compdoc.py", line 398, in locate_named_stream result = self._locate_stream( File "/ragflow/.venv/lib/python3.10/site-packages/xlrd/compdoc.py", line 429, in _locate_stream raise CompDocError("%s corruption: seen[%d] == %d" % (qname, s, self.seen[s])) xlrd.compdoc.CompDocError: Workbook corruption: seen[2] == 4 ``` ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-08-12 12:41:33 +08:00
{ name = "python-calamine", specifier = ">=0.4.0" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "python-docx", specifier = ">=1.1.2,<2.0.0" },
Feat: Gitlab connector (#12248) ### What problem does this PR solve? Feat: Gitlab connector Fix: submit button in darkmode ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2025-12-29 17:05:20 +08:00
{ name = "python-gitlab", specifier = ">=7.0.0" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "python-pptx", specifier = ">=1.0.2,<2.0.0" },
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
{ name = "python-telegram-bot", specifier = ">=21.0" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "qianfan", specifier = "==0.4.6" },
Feat: Alter flask to Quart for async API serving. (#11275) ### What problem does this PR solve? #11277 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-18 17:05:16 +08:00
{ name = "quart-auth", specifier = "==0.11.0" },
{ name = "quart-cors", specifier = "==0.8.0" },
Feat: support openapi (#12521) ### What problem does this PR solve? Support OpenAPI interface description. The issue of not supporting the Swagger interface after upgrading the system framework from Flask to Quart has been resolved. Resolved https://github.com/infiniflow/ragflow/issues/5264 ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: puhaiyang <“761396462@qq.com”>
2026-01-09 17:48:20 +08:00
{ name = "quart-schema", specifier = "==0.23.0" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "ranx", specifier = "==0.3.20" },
Bump python to >=3.12 (#11846) ### What problem does this PR solve? Bump python to >=3.12 ### Type of change - [x] Refactoring
2025-12-09 19:55:25 +08:00
{ name = "readability-lxml", specifier = ">=0.8.4,<1.0.0" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "replicate", specifier = "==0.31.0" },
Feature/docs generator (#11858) ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### What problem does this PR solve? This PR introduces a new Docs Generator agent component for producing downloadable PDF, DOCX, or TXT files from Markdown content generated within a RAGFlow workflow. ### **Key Features** **Backend** - New component: DocsGenerator (agent/component/docs_generator.py) - - Markdown → PDF/DOCX/TXT conversion - - Supports tables, lists, code blocks, headings, and rich formatting - - Configurable document style (fonts, margins, colors, page size, orientation) - - Optional header logo and footer with page numbers/timestamps - **Frontend** - New configuration UI for the Docs Generator - - Download button integrated into the chat interface - - Output wired to the Message component - - Full i18n support **Documentation** Added component guide: docs/guides/agent/agent_component_reference/docs_generator.md **Usage** Add the Docs Generator to a workflow, connect Markdown output from an upstream component, configure metadata/style, and feed its output into the Message component. Users will see a document download button directly in the chat. **Contributor Note** We have been following RAGFlow since more than a year and half now and have worked extensively on personalizing the framework and integrating it into several of our internal systems. Over the past year and a half, we have built multiple platforms that rely on RAGFlow as a core component, which has given us a strong appreciation for how flexible and powerful the project is. We also previously contributed the full Italian translation, and we were glad to see it accepted. This new Docs Generator component was created for our own production needs, and we believe that it may be useful for many others in the community as well. We want to sincerely thank the entire RAGFlow team for the remarkable work you have done and continue to do. If there are opportunities to contribute further, we would be glad to help whenever we have time available. It would be a pleasure to support the project in any way we can. If appropriate, we would be glad to be listed among the project’s contributors, but in any case we look forward to continuing to support and contribute to the project. PentaFrame Development Team --------- Co-authored-by: PentaFrame <info@pentaframe.it> Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2025-12-12 07:59:43 +01:00
{ name = "reportlab", specifier = ">=4.4.1" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "roman-numbers", specifier = "==1.0.2" },
{ name = "ruamel-base", specifier = "==1.0.0" },
{ name = "ruamel-yaml", specifier = ">=0.18.6,<0.19.0" },
{ name = "scholarly", specifier = "==1.7.11" },
{ name = "selenium-wire", specifier = "==5.1.0" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
{ name = "setuptools", specifier = ">=78.1.1,<81.0.0" },
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
{ name = "slack-sdk", specifier = "==3.37.0" },
Remove unused py module dependencies (#11964) ### What problem does this PR solve? ``` pip list --not-required Package Version ---------------------------- --------------------- aiosmtplib 5.0.0 akshare 1.17.94 anthropic 0.34.1 arxiv 2.1.3 Aspose.Slides 24.7.0 atlassian-python-api 4.0.7 azure-identity 1.17.1 azure-storage-file-datalake 12.16.0 bio 1.7.1 boxsdk 10.2.0 captcha 0.7.1 cn2an 0.5.22 cohere 5.6.2 Crawl4AI 0.4.247 dashscope 1.20.11 deepl 1.18.0 demjson3 3.0.6 discord.py 2.3.2 dropbox 12.0.2 duckduckgo_search 7.5.5 editdistance 0.8.1 elasticsearch-dsl 8.12.0 exceptiongroup 1.3.1 extract-msg 0.55.0 ffmpeg-python 0.2.0 flasgger 0.9.7.1 Flask-Cors 5.0.0 Flask-Login 0.6.3 Flask-Mail 0.10.0 Flask-Session 0.8.0 google-auth-oauthlib 1.2.3 google-genai 1.55.0 google-generativeai 0.8.5 google_search_results 2.4.2 graspologic 0.1.dev847+g38e680cab groq 0.9.0 grpcio-status 1.67.1 html_text 0.6.2 imageio-ffmpeg 0.6.0 infinity_emb 0.0.66 infinity-sdk 0.6.11 jira 3.10.5 json_repair 0.35.0 langfuse 3.10.5 mammoth 1.11.0 Markdown 3.6 markdown_to_json 2.1.1 markdownify 1.2.2 mcp 1.19.0 mini-racer 0.12.4 minio 7.2.4 mistralai 0.4.2 moodlepy 0.24.1 mypy-boto3-s3 1.40.26 Office365-REST-Python-Client 2.6.2 ollama 0.6.1 onnxruntime-gpu 1.23.2 opencv-python 4.10.0.84 opencv-python-headless 4.10.0.84 opendal 0.45.20 opensearch-py 2.7.1 ormsgpack 1.5.0 pdfplumber 0.10.4 pip 25.3 pluginlib 0.9.4 psycopg2-binary 2.9.11 pyclipper 1.4.0 pycryptodomex 3.20.0 pyobvector 0.2.18 pyodbc 5.3.0 pypandoc 1.16.2 pypdf 6.4.0 PyPDF2 3.0.1 python-calamine 0.6.1 python-docx 1.2.0 python-pptx 1.0.2 pywencai 0.13.1 qianfan 0.4.6 quart-auth 0.11.0 quart-cors 0.8.0 ranx 0.3.20 readability-lxml 0.8.4.1 replicate 0.31.0 reportlab 4.4.6 roman-numbers 1.0.2 ruamel.base 1.0.0 ruamel.yaml 0.18.16 scholarly 1.7.11 selenium-wire 5.1.0 slack_sdk 3.37.0 socksio 1.0.0 sqlglotrs 0.9.0 StrEnum 0.4.15 tavily-python 0.5.1 tencentcloud-sdk-python 3.0.1478 tika 2.6.0 valkey 6.0.2 vertexai 1.70.0 volcengine 1.0.194 voyageai 0.2.3 webdav4 0.10.0 webdriver-manager 4.0.1 wikipedia 1.4.0 word2number 1.1 xgboost 1.6.0 xpinyin 0.7.6 yfinance 0.2.65 zhipuai 2.0.1 ``` ### Type of change - [x] Refactoring Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-16 12:40:03 +08:00
{ name = "socksio", specifier = "==1.0.0" },
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
{ name = "spacy", specifier = "==3.8.14" },
Remove unused py module dependencies (#11964) ### What problem does this PR solve? ``` pip list --not-required Package Version ---------------------------- --------------------- aiosmtplib 5.0.0 akshare 1.17.94 anthropic 0.34.1 arxiv 2.1.3 Aspose.Slides 24.7.0 atlassian-python-api 4.0.7 azure-identity 1.17.1 azure-storage-file-datalake 12.16.0 bio 1.7.1 boxsdk 10.2.0 captcha 0.7.1 cn2an 0.5.22 cohere 5.6.2 Crawl4AI 0.4.247 dashscope 1.20.11 deepl 1.18.0 demjson3 3.0.6 discord.py 2.3.2 dropbox 12.0.2 duckduckgo_search 7.5.5 editdistance 0.8.1 elasticsearch-dsl 8.12.0 exceptiongroup 1.3.1 extract-msg 0.55.0 ffmpeg-python 0.2.0 flasgger 0.9.7.1 Flask-Cors 5.0.0 Flask-Login 0.6.3 Flask-Mail 0.10.0 Flask-Session 0.8.0 google-auth-oauthlib 1.2.3 google-genai 1.55.0 google-generativeai 0.8.5 google_search_results 2.4.2 graspologic 0.1.dev847+g38e680cab groq 0.9.0 grpcio-status 1.67.1 html_text 0.6.2 imageio-ffmpeg 0.6.0 infinity_emb 0.0.66 infinity-sdk 0.6.11 jira 3.10.5 json_repair 0.35.0 langfuse 3.10.5 mammoth 1.11.0 Markdown 3.6 markdown_to_json 2.1.1 markdownify 1.2.2 mcp 1.19.0 mini-racer 0.12.4 minio 7.2.4 mistralai 0.4.2 moodlepy 0.24.1 mypy-boto3-s3 1.40.26 Office365-REST-Python-Client 2.6.2 ollama 0.6.1 onnxruntime-gpu 1.23.2 opencv-python 4.10.0.84 opencv-python-headless 4.10.0.84 opendal 0.45.20 opensearch-py 2.7.1 ormsgpack 1.5.0 pdfplumber 0.10.4 pip 25.3 pluginlib 0.9.4 psycopg2-binary 2.9.11 pyclipper 1.4.0 pycryptodomex 3.20.0 pyobvector 0.2.18 pyodbc 5.3.0 pypandoc 1.16.2 pypdf 6.4.0 PyPDF2 3.0.1 python-calamine 0.6.1 python-docx 1.2.0 python-pptx 1.0.2 pywencai 0.13.1 qianfan 0.4.6 quart-auth 0.11.0 quart-cors 0.8.0 ranx 0.3.20 readability-lxml 0.8.4.1 replicate 0.31.0 reportlab 4.4.6 roman-numbers 1.0.2 ruamel.base 1.0.0 ruamel.yaml 0.18.16 scholarly 1.7.11 selenium-wire 5.1.0 slack_sdk 3.37.0 socksio 1.0.0 sqlglotrs 0.9.0 StrEnum 0.4.15 tavily-python 0.5.1 tencentcloud-sdk-python 3.0.1478 tika 2.6.0 valkey 6.0.2 vertexai 1.70.0 volcengine 1.0.194 voyageai 0.2.3 webdav4 0.10.0 webdriver-manager 4.0.1 wikipedia 1.4.0 word2number 1.1 xgboost 1.6.0 xpinyin 0.7.6 yfinance 0.2.65 zhipuai 2.0.1 ``` ### Type of change - [x] Refactoring Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-16 12:40:03 +08:00
{ name = "sqlglotrs", specifier = "==0.9.0" },
Add tavily as web searh tool. (#5349) ### What problem does this PR solve? #5198 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-02-26 10:21:04 +08:00
{ name = "tavily-python", specifier = "==0.5.1" },
Feat: add TCADP Parser (#10775) ### What problem does this PR solve? This PR adds a new TCADP (Tencent Cloud Advanced Document Processing) parser to RAGFlow, enabling users to leverage Tencent Cloud's document parsing capabilities for more accurate and structured document processing. The implementation includes: New TCADP Parser: A complete implementation of Tencent Cloud's document parsing API without SDK dependency Configuration Support: Added configuration options in service_conf.yaml for Tencent Cloud API credentials Frontend Integration: Updated UI components to support the new TCADP parser option Error Handling: Comprehensive error handling and retry mechanisms for API calls Result Processing: Support for both SSE streaming and JSON response formats from Tencent Cloud API ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2025-10-27 15:14:58 +08:00
{ name = "tencentcloud-sdk-python", specifier = "==3.0.1478" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "tika", specifier = "==2.6.0" },
{ name = "valkey", specifier = "==6.0.2" },
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
{ name = "volcengine", specifier = "==1.0.194" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "voyageai", specifier = "==0.2.3" },
Feat: Add Webdav storage as data source (#11422) ### What problem does this PR solve? This PR adds webdav storage as data source for data sync service. ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-26 07:14:42 +01:00
{ name = "webdav4", specifier = ">=0.10.0,<0.11.0" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "webdriver-manager", specifier = "==4.0.1" },
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
{ name = "wechatpy", specifier = ">=1.8.18" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "wikipedia", specifier = "==1.4.0" },
{ name = "word2number", specifier = "==1.1" },
Fix: revert xgboost version to 1.6.0 (#13984) ### What problem does this PR solve? Revert xgboost version to 1.6.0 ### Type of change - [ ] Bug Fix (non-breaking change which fixes an issue) - [ ] New Feature (non-breaking change which adds functionality) - [ ] Documentation Update - [ ] Refactoring - [ ] Performance Improvement - [ ] Other (please describe): <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Chores** * Updated xgboost dependency from version 3.2.0 to 1.6.0 <!-- end of auto-generated comment: release notes by coderabbit.ai -->
2026-04-08 19:53:47 +08:00
{ name = "xgboost", specifier = "==1.6.0" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "xpinyin", specifier = "==0.7.6" },
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
{ name = "yfinance", specifier = "==0.2.65" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "zhipuai", specifier = "==2.0.1" },
]
Feat: sync test group to top pyproject.toml (#8015) ### What problem does this PR solve? sync test group from sdk/python/pyproject.toml to top pyproject.toml ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-06-03 15:21:06 +08:00
[package.metadata.requires-dev]
test = [
Test: Add code coverage reporting to CI (#12874) ### What problem does this PR solve? Add code coverage reporting to CI ### Type of change - [x] Test (please describe): coverage report --------- Co-authored-by: Liu An <asiro@qq.com>
2026-01-30 14:49:16 +08:00
{ name = "codecov", specifier = ">=2.1.13" },
Feat: sync test group to top pyproject.toml (#8015) ### What problem does this PR solve? sync test group from sdk/python/pyproject.toml to top pyproject.toml ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-06-03 15:21:06 +08:00
{ name = "hypothesis", specifier = ">=6.132.0" },
{ name = "openpyxl", specifier = ">=3.1.5" },
Chore(deps-dev): Bump pillow from 12.1.1 to 12.2.0 (#14578) As title
2026-05-06 11:08:38 +08:00
{ name = "pillow", specifier = ">=12.2.0,<13.0.0" },
Feat: Benchmark CLI additions and documentation (#12536) ### What problem does this PR solve? This PR adds a dedicated HTTP benchmark CLI for RAGFlow chat and retrieval endpoints so we can measure latency/QPS. ### Type of change - [x] Documentation Update - [x] Other (please describe): Adds a CLI benchmarking tool for chat/retrieval latency/QPS --------- Co-authored-by: Liu An <asiro@qq.com>
2026-01-14 13:49:16 +08:00
{ name = "pycryptodomex", specifier = "==3.20.0" },
Feat: sync test group to top pyproject.toml (#8015) ### What problem does this PR solve? sync test group from sdk/python/pyproject.toml to top pyproject.toml ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-06-03 15:21:06 +08:00
{ name = "pytest", specifier = ">=8.3.5" },
Add ut test into CI (#11753) ### What problem does this PR solve? As title ### Type of change - [x] Other (please describe): --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-05 11:40:16 +08:00
{ name = "pytest-asyncio", specifier = ">=1.3.0" },
{ name = "pytest-cov", specifier = ">=7.0.0" },
Update graspologic to gitee (#13362) ### What problem does this PR solve? Accelerate python module downloading ### Type of change - [x] Refactoring Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2026-03-04 17:48:47 +08:00
{ name = "pytest-playwright", specifier = ">=0.7.2" },
Add ut test into CI (#11753) ### What problem does this PR solve? As title ### Type of change - [x] Other (please describe): --------- Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2025-12-05 11:40:16 +08:00
{ name = "pytest-xdist", specifier = ">=3.8.0" },
Feat: sync test group to top pyproject.toml (#8015) ### What problem does this PR solve? sync test group from sdk/python/pyproject.toml to top pyproject.toml ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-06-03 15:21:06 +08:00
{ name = "python-docx", specifier = ">=1.1.2" },
{ name = "python-pptx", specifier = ">=1.0.2" },
{ name = "reportlab", specifier = ">=4.4.1" },
{ name = "requests", specifier = ">=2.32.2" },
{ name = "requests-toolbelt", specifier = ">=1.0.0" },
]
Refa: migrate chat models to LiteLLM (#9394) ### What problem does this PR solve? All models pass the mock response tests, which means that if a model can return the correct response, everything should work as expected. However, not all models have been fully tested in a real environment, the real API_KEY. I suggest actively monitoring the refactored models over the coming period to ensure they work correctly and fixing them step by step, or waiting to merge until most have been tested in practical environment. ### Type of change - [x] Refactoring
2025-08-12 10:59:20 +08:00
[[package]]
name = "rank-bm25"
version = "0.2.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Refa: migrate chat models to LiteLLM (#9394) ### What problem does this PR solve? All models pass the mock response tests, which means that if a model can return the correct response, everything should work as expected. However, not all models have been fully tested in a real environment, the real API_KEY. I suggest actively monitoring the refactored models over the coming period to ensure they work correctly and fixing them step by step, or waiting to merge until most have been tested in practical environment. ### Type of change - [x] Refactoring
2025-08-12 10:59:20 +08:00
dependencies = [
{ name = "numpy" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/fc/0a/f9579384aa017d8b4c15613f86954b92a95a93d641cc849182467cf0bb3b/rank_bm25-0.2.2.tar.gz", hash = "sha256:096ccef76f8188563419aaf384a02f0ea459503fdf77901378d4fd9d87e5e51d" }
Refa: migrate chat models to LiteLLM (#9394) ### What problem does this PR solve? All models pass the mock response tests, which means that if a model can return the correct response, everything should work as expected. However, not all models have been fully tested in a real environment, the real API_KEY. I suggest actively monitoring the refactored models over the coming period to ensure they work correctly and fixing them step by step, or waiting to merge until most have been tested in practical environment. ### Type of change - [x] Refactoring
2025-08-12 10:59:20 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/2a/21/f691fb2613100a62b3fa91e9988c991e9ca5b89ea31c0d3152a3210344f9/rank_bm25-0.2.2-py3-none-any.whl", hash = "sha256:7bd4a95571adadfc271746fa146a4bcfd89c0cf731e49c3d1ad863290adbe8ae" },
Refa: migrate chat models to LiteLLM (#9394) ### What problem does this PR solve? All models pass the mock response tests, which means that if a model can return the correct response, everything should work as expected. However, not all models have been fully tested in a real environment, the real API_KEY. I suggest actively monitoring the refactored models over the coming period to ensure they work correctly and fixing them step by step, or waiting to merge until most have been tested in practical environment. ### Type of change - [x] Refactoring
2025-08-12 10:59:20 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "ranx"
version = "0.3.20"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "cbor2" },
{ name = "fastparquet" },
{ name = "ir-datasets" },
{ name = "lz4" },
{ name = "numba" },
{ name = "numpy" },
{ name = "orjson" },
{ name = "pandas" },
{ name = "rich" },
Bump infinity to v0.6.11. Requires python>=3.11 (#11814) ### What problem does this PR solve? Bump infinity to v0.6.11. Requires python>=3.11 ### Type of change - [x] Refactoring
2025-12-09 16:23:37 +08:00
{ name = "scipy" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "seaborn" },
{ name = "tabulate" },
{ name = "tqdm" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/35/fe/4d4e7c69137afdeb5a4a85afcf04b84f087a284b7f22034e2e13e121de83/ranx-0.3.20.tar.gz", hash = "sha256:8afc6f2042c40645e5d1fd80c35ed75a885e18bd2db7e95cc7ec32a0b41e59ea" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/1e/30/53f41b7b728a48da8974075f56c57200d7b11e4e9fa93be3cabf8218dc0c/ranx-0.3.20-py3-none-any.whl", hash = "sha256:e056e4d5981b0328b045868cc7064fc57a545f36009fbe9bb602295ec33335de" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "readability-lxml"
Bump python to >=3.12 (#11846) ### What problem does this PR solve? Bump python to >=3.12 ### Type of change - [x] Refactoring
2025-12-09 19:55:25 +08:00
version = "0.8.4.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "chardet" },
{ name = "cssselect" },
Bump python to >=3.12 (#11846) ### What problem does this PR solve? Bump python to >=3.12 ### Type of change - [x] Refactoring
2025-12-09 19:55:25 +08:00
{ name = "lxml", extra = ["html-clean"] },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/55/3e/dc87d97532ddad58af786ec89c7036182e352574c1cba37bf2bf783d2b15/readability_lxml-0.8.4.1.tar.gz", hash = "sha256:9d2924f5942dd7f37fb4da353263b22a3e877ccf922d0e45e348e4177b035a53" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/c7/75/2cc58965097e351415af420be81c4665cf80da52a17ef43c01ffbe2caf91/readability_lxml-0.8.4.1-py3-none-any.whl", hash = "sha256:874c0cea22c3bf2b78c7f8df831bfaad3c0a89b7301d45a188db581652b4b465" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "readerwriterlock"
version = "1.0.9"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "typing-extensions" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a4/b9/6b7c390440ec23bf5fdf33e76d6c3b697a788b983c11cb2739d6541835d6/readerwriterlock-1.0.9.tar.gz", hash = "sha256:b7c4cc003435d7a8ff15b312b0a62a88d9800ba6164af88991f87f8b748f9bea" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/c2/5a/2f2e7fc026d5e64b5408aa3fbe0296a6407b8481196cae4daacacb3a3ae0/readerwriterlock-1.0.9-py3-none-any.whl", hash = "sha256:8c4b704e60d15991462081a27ef46762fea49b478aa4426644f2146754759ca7" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
[[package]]
name = "red-black-tree-mod"
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
version = "1.22"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/48/75/bfa342a2ebfc9623b701f1c6995b9906fd6dd2cedf6bce777d09e23303ac/red-black-tree-mod-1.22.tar.gz", hash = "sha256:38e3652903a2bf96379c27c2082ca0b7b905158662dd7ef0c97f4fd93a9aa908" }
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "referencing"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "0.37.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "attrs" },
{ name = "rpds-py" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/22/f5/df4e9027acead3ecc63e50fe1e36aca1523e1719559c499951bb4b53188f/referencing-0.37.0.tar.gz", hash = "sha256:44aefc3142c5b842538163acb373e24cce6632bd54bdb01b21ad5863489f50d8" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/2c/58/ca301544e1fa93ed4f80d724bf5b194f6e4b945841c5bfd555878eea9fcb/referencing-0.37.0-py3-none-any.whl", hash = "sha256:381329a9f99628c9069361716891d34ad94af76e461dcb0335825aecc7692231" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "regex"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "2025.11.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/cc/a9/546676f25e573a4cf00fe8e119b78a37b6a8fe2dc95cda877b30889c9c45/regex-2025.11.3.tar.gz", hash = "sha256:1fedc720f9bb2494ce31a58a1631f9c82df6a09b49c19517ea5cc280b4541e01" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/e1/a7/dda24ebd49da46a197436ad96378f17df30ceb40e52e859fc42cac45b850/regex-2025.11.3-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:c1e448051717a334891f2b9a620fe36776ebf3dd8ec46a0b877c8ae69575feb4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/19/22/af2dc751aacf88089836aa088a1a11c4f21a04707eb1b0478e8e8fb32847/regex-2025.11.3-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:9b5aca4d5dfd7fbfbfbdaf44850fcc7709a01146a797536a8f84952e940cca76" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a3/88/1a3ea5672f4b0a84802ee9891b86743438e7c04eb0b8f8c4e16a42375327/regex-2025.11.3-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:04d2765516395cf7dda331a244a3282c0f5ae96075f728629287dfa6f76ba70a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fb/8c/f5987895bf42b8ddeea1b315c9fedcfe07cadee28b9c98cf50d00adcb14d/regex-2025.11.3-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:5d9903ca42bfeec4cebedba8022a7c97ad2aab22e09573ce9976ba01b65e4361" },
{ url = "https://mirrors.aliyun.com/pypi/packages/99/2a/6591ebeede78203fa77ee46a1c36649e02df9eaa77a033d1ccdf2fcd5d4e/regex-2025.11.3-cp313-cp313-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:639431bdc89d6429f6721625e8129413980ccd62e9d3f496be618a41d205f160" },
{ url = "https://mirrors.aliyun.com/pypi/packages/94/d6/be32a87cf28cf8ed064ff281cfbd49aefd90242a83e4b08b5a86b38e8eb4/regex-2025.11.3-cp313-cp313-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:f117efad42068f9715677c8523ed2be1518116d1c49b1dd17987716695181efe" },
{ url = "https://mirrors.aliyun.com/pypi/packages/62/11/9bcef2d1445665b180ac7f230406ad80671f0fc2a6ffb93493b5dd8cd64c/regex-2025.11.3-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:4aecb6f461316adf9f1f0f6a4a1a3d79e045f9b71ec76055a791affa3b285850" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e5/a7/da0dc273d57f560399aa16d8a68ae7f9b57679476fc7ace46501d455fe84/regex-2025.11.3-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:3b3a5f320136873cc5561098dfab677eea139521cb9a9e8db98b7e64aef44cbc" },
{ url = "https://mirrors.aliyun.com/pypi/packages/da/4b/732a0c5a9736a0b8d6d720d4945a2f1e6f38f87f48f3173559f53e8d5d82/regex-2025.11.3-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:75fa6f0056e7efb1f42a1c34e58be24072cb9e61a601340cc1196ae92326a4f9" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0c/f5/a2a03df27dc4c2d0c769220f5110ba8c4084b0bfa9ab0f9b4fcfa3d2b0fc/regex-2025.11.3-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:dbe6095001465294f13f1adcd3311e50dd84e5a71525f20a10bd16689c61ce0b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d6/09/e1cd5bee3841c7f6eb37d95ca91cdee7100b8f88b81e41c2ef426910891a/regex-2025.11.3-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:454d9b4ae7881afbc25015b8627c16d88a597479b9dea82b8c6e7e2e07240dc7" },
{ url = "https://mirrors.aliyun.com/pypi/packages/eb/51/702f5ea74e2a9c13d855a6a85b7f80c30f9e72a95493260193c07f3f8d74/regex-2025.11.3-cp313-cp313-win32.whl", hash = "sha256:28ba4d69171fc6e9896337d4fc63a43660002b7da53fc15ac992abcf3410917c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8b/00/6e29bb314e271a743170e53649db0fdb8e8ff0b64b4f425f5602f4eb9014/regex-2025.11.3-cp313-cp313-win_amd64.whl", hash = "sha256:bac4200befe50c670c405dc33af26dad5a3b6b255dd6c000d92fe4629f9ed6a5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/25/f1/b156ff9f2ec9ac441710764dda95e4edaf5f36aca48246d1eea3f1fd96ec/regex-2025.11.3-cp313-cp313-win_arm64.whl", hash = "sha256:2292cd5a90dab247f9abe892ac584cb24f0f54680c73fcb4a7493c66c2bf2467" },
{ url = "https://mirrors.aliyun.com/pypi/packages/20/28/fd0c63357caefe5680b8ea052131acbd7f456893b69cc2a90cc3e0dc90d4/regex-2025.11.3-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:1eb1ebf6822b756c723e09f5186473d93236c06c579d2cc0671a722d2ab14281" },
{ url = "https://mirrors.aliyun.com/pypi/packages/df/ec/7014c15626ab46b902b3bcc4b28a7bae46d8f281fc7ea9c95e22fcaaa917/regex-2025.11.3-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:1e00ec2970aab10dc5db34af535f21fcf32b4a31d99e34963419636e2f85ae39" },
{ url = "https://mirrors.aliyun.com/pypi/packages/23/ab/3b952ff7239f20d05f1f99e9e20188513905f218c81d52fb5e78d2bf7634/regex-2025.11.3-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:a4cb042b615245d5ff9b3794f56be4138b5adc35a4166014d31d1814744148c7" },
{ url = "https://mirrors.aliyun.com/pypi/packages/21/7e/3dc2749fc684f455f162dcafb8a187b559e2614f3826877d3844a131f37b/regex-2025.11.3-cp313-cp313t-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:44f264d4bf02f3176467d90b294d59bf1db9fe53c141ff772f27a8b456b2a9ed" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1b/0b/d529a85ab349c6a25d1ca783235b6e3eedf187247eab536797021f7126c6/regex-2025.11.3-cp313-cp313t-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:7be0277469bf3bd7a34a9c57c1b6a724532a0d235cd0dc4e7f4316f982c28b19" },
{ url = "https://mirrors.aliyun.com/pypi/packages/7d/18/2d868155f8c9e3e9d8f9e10c64e9a9f496bb8f7e037a88a8bed26b435af6/regex-2025.11.3-cp313-cp313t-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:0d31e08426ff4b5b650f68839f5af51a92a5b51abd8554a60c2fbc7c71f25d0b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2d/71/9d72ff0f354fa783fe2ba913c8734c3b433b86406117a8db4ea2bf1c7a2f/regex-2025.11.3-cp313-cp313t-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:e43586ce5bd28f9f285a6e729466841368c4a0353f6fd08d4ce4630843d3648a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e7/19/ce4bf7f5575c97f82b6e804ffb5c4e940c62609ab2a0d9538d47a7fdf7d4/regex-2025.11.3-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:0f9397d561a4c16829d4e6ff75202c1c08b68a3bdbfe29dbfcdb31c9830907c6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/03/86/fd1063a176ffb7b2315f9a1b08d17b18118b28d9df163132615b835a26ee/regex-2025.11.3-cp313-cp313t-musllinux_1_2_ppc64le.whl", hash = "sha256:dd16e78eb18ffdb25ee33a0682d17912e8cc8a770e885aeee95020046128f1ce" },
{ url = "https://mirrors.aliyun.com/pypi/packages/12/43/103fb2e9811205e7386366501bc866a164a0430c79dd59eac886a2822950/regex-2025.11.3-cp313-cp313t-musllinux_1_2_s390x.whl", hash = "sha256:ffcca5b9efe948ba0661e9df0fa50d2bc4b097c70b9810212d6b62f05d83b2dd" },
{ url = "https://mirrors.aliyun.com/pypi/packages/7d/22/e392e53f3869b75804762c7c848bd2dd2abf2b70fb0e526f58724638bd35/regex-2025.11.3-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:c56b4d162ca2b43318ac671c65bd4d563e841a694ac70e1a976ac38fcf4ca1d2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4f/f9/8bd6b656592f925b6845fcbb4d57603a3ac2fb2373344ffa1ed70aa6820a/regex-2025.11.3-cp313-cp313t-win32.whl", hash = "sha256:9ddc42e68114e161e51e272f667d640f97e84a2b9ef14b7477c53aac20c2d59a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e5/87/0e7d603467775ff65cd2aeabf1b5b50cc1c3708556a8b849a2fa4dd1542b/regex-2025.11.3-cp313-cp313t-win_amd64.whl", hash = "sha256:7a7c7fdf755032ffdd72c77e3d8096bdcb0eb92e89e17571a196f03d88b11b3c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8d/d0/2afc6f8e94e2b64bfb738a7c2b6387ac1699f09f032d363ed9447fd2bb57/regex-2025.11.3-cp313-cp313t-win_arm64.whl", hash = "sha256:df9eb838c44f570283712e7cff14c16329a9f0fb19ca492d21d4b7528ee6821e" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "replicate"
version = "0.31.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "httpx" },
{ name = "packaging" },
{ name = "pydantic" },
{ name = "typing-extensions" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/02/eb/04fbc1787d4d688feafa080b11e6672e819d170d9fda9ae4a2c2ac1e3dc2/replicate-0.31.0.tar.gz", hash = "sha256:6503f5266e08f7bd0f125f735a7dd68a298496b9f057be0f101aa7e8c7280728" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/77/0f/f6067b7076faee22aef6190f703524e8ba8eac490191352c5cb0253c4823/replicate-0.31.0-py3-none-any.whl", hash = "sha256:27ee067ccb4c37d8c2fc5ab87bb312da36447dfcd12527002bbd0b78f6ef195a" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: sync test group to top pyproject.toml (#8015) ### What problem does this PR solve? sync test group from sdk/python/pyproject.toml to top pyproject.toml ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-06-03 15:21:06 +08:00
[[package]]
name = "reportlab"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "4.4.7"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: sync test group to top pyproject.toml (#8015) ### What problem does this PR solve? sync test group from sdk/python/pyproject.toml to top pyproject.toml ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-06-03 15:21:06 +08:00
dependencies = [
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
{ name = "charset-normalizer" },
Feat: sync test group to top pyproject.toml (#8015) ### What problem does this PR solve? sync test group from sdk/python/pyproject.toml to top pyproject.toml ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-06-03 15:21:06 +08:00
{ name = "pillow" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f8/a7/4600cb1cfc975a06552e8927844ddcb8fd90217e9a6068f5c7aa76c3f221/reportlab-4.4.7.tar.gz", hash = "sha256:41e8287af965e5996764933f3e75e7f363c3b6f252ba172f9429e81658d7b170" }
Feat: sync test group to top pyproject.toml (#8015) ### What problem does this PR solve? sync test group from sdk/python/pyproject.toml to top pyproject.toml ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-06-03 15:21:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/e7/bf/a29507386366ab17306b187ad247dd78e4599be9032cb5f44c940f547fc0/reportlab-4.4.7-py3-none-any.whl", hash = "sha256:8fa05cbf468e0e76745caf2029a4770276edb3c8e86a0b71e0398926baf50673" },
Feat: sync test group to top pyproject.toml (#8015) ### What problem does this PR solve? sync test group from sdk/python/pyproject.toml to top pyproject.toml ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-06-03 15:21:06 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "requests"
Bump infinity to v0.6.11. Requires python>=3.11 (#11814) ### What problem does this PR solve? Bump infinity to v0.6.11. Requires python>=3.11 ### Type of change - [x] Refactoring
2025-12-09 16:23:37 +08:00
version = "2.32.5"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "certifi" },
{ name = "charset-normalizer" },
{ name = "idna" },
{ name = "urllib3" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/c9/74/b3ff8e6c8446842c3f5c837e9c3dfcfe2018ea6ecef224c710c85ef728f4/requests-2.32.5.tar.gz", hash = "sha256:dbba0bac56e100853db0ea71b82b4dfd5fe2bf6d3754a8893c3af500cec7d7cf" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/1e/db/4254e3eabe8020b458f1a747140d32277ec7a271daf1d235b70dc0b4e6e3/requests-2.32.5-py3-none-any.whl", hash = "sha256:2462f94637a34fd532264295e186976db0f5d453d1cdd31473c85a6a161affb6" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[package.optional-dependencies]
socks = [
{ name = "pysocks" },
]
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
[[package]]
name = "requests-oauthlib"
version = "2.0.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
dependencies = [
{ name = "oauthlib" },
{ name = "requests" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/42/f2/05f29bc3913aea15eb670be136045bf5c5bbf4b99ecb839da9b422bb2c85/requests-oauthlib-2.0.0.tar.gz", hash = "sha256:b3dffaebd884d8cd778494369603a9e7b58d29111bf6b41bdc2dcd87203af4e9" }
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/3b/5d/63d4ae3b9daea098d5d6f5da83984853c1bbacd5dc826764b249fe119d24/requests_oauthlib-2.0.0-py2.py3-none-any.whl", hash = "sha256:7dd8a5c40426b779b0868c404bdef9768deccf22749cde15852df527e6269b36" },
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
]
Feat: sync test group to top pyproject.toml (#8015) ### What problem does this PR solve? sync test group from sdk/python/pyproject.toml to top pyproject.toml ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-06-03 15:21:06 +08:00
[[package]]
name = "requests-toolbelt"
version = "1.0.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: sync test group to top pyproject.toml (#8015) ### What problem does this PR solve? sync test group from sdk/python/pyproject.toml to top pyproject.toml ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-06-03 15:21:06 +08:00
dependencies = [
{ name = "requests" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f3/61/d7545dafb7ac2230c70d38d31cbfe4cc64f7144dc41f6e4e4b78ecd9f5bb/requests-toolbelt-1.0.0.tar.gz", hash = "sha256:7681a0a3d047012b5bdc0ee37d7f8f07ebe76ab08caeccfc3921ce23c88d5bc6" }
Feat: sync test group to top pyproject.toml (#8015) ### What problem does this PR solve? sync test group from sdk/python/pyproject.toml to top pyproject.toml ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-06-03 15:21:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/3f/51/d4db610ef29373b879047326cbf6fa98b6c1969d6f6dc423279de2b1be2c/requests_toolbelt-1.0.0-py2.py3-none-any.whl", hash = "sha256:cccfdd665f0a24fcf4726e690f65639d272bb0637b9b92dfd91a5568ccf6bd06" },
Feat: sync test group to top pyproject.toml (#8015) ### What problem does this PR solve? sync test group from sdk/python/pyproject.toml to top pyproject.toml ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-06-03 15:21:06 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "retry"
version = "0.9.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "decorator" },
{ name = "py" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/9d/72/75d0b85443fbc8d9f38d08d2b1b67cc184ce35280e4a3813cda2f445f3a4/retry-0.9.2.tar.gz", hash = "sha256:f8bfa8b99b69c4506d6f5bd3b0aabf77f98cdb17f3c9fc3f5ca820033336fba4" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/4b/0d/53aea75710af4528a25ed6837d71d117602b01946b307a3912cb3cfcbcba/retry-0.9.2-py2.py3-none-any.whl", hash = "sha256:ccddf89761fa2c726ab29391837d4327f819ea14d244c232a1d24c67a2f98606" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "rich"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "14.2.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "markdown-it-py" },
{ name = "pygments" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/fb/d2/8920e102050a0de7bfabeb4c4614a49248cf8d5d7a8d01885fbb24dc767a/rich-14.2.0.tar.gz", hash = "sha256:73ff50c7c0c1c77c8243079283f4edb376f0f6442433aecb8ce7e6d0b92d1fe4" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/25/7a/b0178788f8dc6cafce37a212c99565fa1fe7872c70c6c9c1e1a372d9d88f/rich-14.2.0-py3-none-any.whl", hash = "sha256:76bc51fe2e57d2b1be1f96c524b890b816e334ab4c1e45888799bfaab0021edd" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "roman-numbers"
version = "1.0.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/95/ce/e9f6b0d260f48713f2d735e0986ee4ead311cd168c217c5f94b0fad6817b/roman_numbers-1.0.2.tar.gz", hash = "sha256:fb84b7755ba972d549e73fac1c100f0eeb9fc247474d43d0f433c0b72152c699" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/9a/85/09e9e6bd6cd4cc0ed463d2b6ce3c7741698d45ca157318730a1346df4819/roman_numbers-1.0.2-py3-none-any.whl", hash = "sha256:ffbc00aaf41538208f975d1b1ccfe80372bae1866e7cd632862d8c6b45edf447" },
Add tavily as web searh tool. (#5349) ### What problem does this PR solve? #5198 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-02-26 10:21:04 +08:00
]
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
[[package]]
name = "roman-numerals"
version = "4.1.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ae/f9/41dc953bbeb056c17d5f7a519f50fdf010bd0553be2d630bc69d1e022703/roman_numerals-4.1.0.tar.gz", hash = "sha256:1af8b147eb1405d5839e78aeb93131690495fe9da5c91856cb33ad55a7f1e5b2" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/04/54/6f679c435d28e0a568d8e8a7c0a93a09010818634c3c3907fc98d8983770/roman_numerals-4.1.0-py3-none-any.whl", hash = "sha256:647ba99caddc2cc1e55a51e4360689115551bf4476d90e8162cf8c345fe233c7" },
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
]
[[package]]
name = "roman-numerals-py"
version = "4.1.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
dependencies = [
{ name = "roman-numerals" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/cb/b5/de96fca640f4f656eb79bbee0e79aeec52e3e0e359f8a3e6a0d366378b64/roman_numerals_py-4.1.0.tar.gz", hash = "sha256:f5d7b2b4ca52dd855ef7ab8eb3590f428c0b1ea480736ce32b01fef2a5f8daf9" }
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/27/2c/daca29684cbe9fd4bc711f8246da3c10adca1ccc4d24436b17572eb2590e/roman_numerals_py-4.1.0-py3-none-any.whl", hash = "sha256:553114c1167141c1283a51743759723ecd05604a1b6b507225e91dc1a6df0780" },
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "rpds-py"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "0.30.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/20/af/3f2f423103f1113b36230496629986e0ef7e199d2aa8392452b484b38ced/rpds_py-0.30.0.tar.gz", hash = "sha256:dd8ff7cf90014af0c0f787eea34794ebf6415242ee1d6fa91eaba725cc441e84" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/ed/dc/d61221eb88ff410de3c49143407f6f3147acf2538c86f2ab7ce65ae7d5f9/rpds_py-0.30.0-cp313-cp313-macosx_10_12_x86_64.whl", hash = "sha256:f83424d738204d9770830d35290ff3273fbb02b41f919870479fab14b9d303b2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fd/32/55fb50ae104061dbc564ef15cc43c013dc4a9f4527a1f4d99baddf56fe5f/rpds_py-0.30.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:e7536cd91353c5273434b4e003cbda89034d67e7710eab8761fd918ec6c69cf8" },
{ url = "https://mirrors.aliyun.com/pypi/packages/58/70/faed8186300e3b9bdd138d0273109784eea2396c68458ed580f885dfe7ad/rpds_py-0.30.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2771c6c15973347f50fece41fc447c054b7ac2ae0502388ce3b6738cd366e3d4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/bd/a8/073cac3ed2c6387df38f71296d002ab43496a96b92c823e76f46b8af0543/rpds_py-0.30.0-cp313-cp313-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:0a59119fc6e3f460315fe9d08149f8102aa322299deaa5cab5b40092345c2136" },
{ url = "https://mirrors.aliyun.com/pypi/packages/77/57/5999eb8c58671f1c11eba084115e77a8899d6e694d2a18f69f0ba471ec8b/rpds_py-0.30.0-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:76fec018282b4ead0364022e3c54b60bf368b9d926877957a8624b58419169b7" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e0/af/5ab4833eadc36c0a8ed2bc5c0de0493c04f6c06de223170bd0798ff98ced/rpds_py-0.30.0-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:692bef75a5525db97318e8cd061542b5a79812d711ea03dbc1f6f8dbb0c5f0d2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b7/de/f7192e12b21b9e9a68a6d0f249b4af3fdcdff8418be0767a627564afa1f1/rpds_py-0.30.0-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9027da1ce107104c50c81383cae773ef5c24d296dd11c99e2629dbd7967a20c6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/91/c4/fc70cd0249496493500e7cc2de87504f5aa6509de1e88623431fec76d4b6/rpds_py-0.30.0-cp313-cp313-manylinux_2_31_riscv64.whl", hash = "sha256:9cf69cdda1f5968a30a359aba2f7f9aa648a9ce4b580d6826437f2b291cfc86e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/58/95/d9275b05ab96556fefff73a385813eb66032e4c99f411d0795372d9abcea/rpds_py-0.30.0-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:a4796a717bf12b9da9d3ad002519a86063dcac8988b030e405704ef7d74d2d9d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/06/c1/3088fc04b6624eb12a57eb814f0d4997a44b0d208d6cace713033ff1a6ba/rpds_py-0.30.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:5d4c2aa7c50ad4728a094ebd5eb46c452e9cb7edbfdb18f9e1221f597a73e1e7" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d8/42/c612a833183b39774e8ac8fecae81263a68b9583ee343db33ab571a7ce55/rpds_py-0.30.0-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:ba81a9203d07805435eb06f536d95a266c21e5b2dfbf6517748ca40c98d19e31" },
{ url = "https://mirrors.aliyun.com/pypi/packages/5f/60/525a50f45b01d70005403ae0e25f43c0384369ad24ffe46e8d9068b50086/rpds_py-0.30.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:945dccface01af02675628334f7cf49c2af4c1c904748efc5cf7bbdf0b579f95" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0b/5d/47c4655e9bcd5ca907148535c10e7d489044243cc9941c16ed7cd53be91d/rpds_py-0.30.0-cp313-cp313-win32.whl", hash = "sha256:b40fb160a2db369a194cb27943582b38f79fc4887291417685f3ad693c5a1d5d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f2/e1/485132437d20aa4d3e1d8b3fb5a5e65aa8139f1e097080c2a8443201742c/rpds_py-0.30.0-cp313-cp313-win_amd64.whl", hash = "sha256:806f36b1b605e2d6a72716f321f20036b9489d29c51c91f4dd29a3e3afb73b15" },
{ url = "https://mirrors.aliyun.com/pypi/packages/24/95/ffd128ed1146a153d928617b0ef673960130be0009c77d8fbf0abe306713/rpds_py-0.30.0-cp313-cp313-win_arm64.whl", hash = "sha256:d96c2086587c7c30d44f31f42eae4eac89b60dabbac18c7669be3700f13c3ce1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ff/1b/b10de890a0def2a319a2626334a7f0ae388215eb60914dbac8a3bae54435/rpds_py-0.30.0-cp313-cp313t-macosx_10_12_x86_64.whl", hash = "sha256:eb0b93f2e5c2189ee831ee43f156ed34e2a89a78a66b98cadad955972548be5a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0d/bf/27e39f5971dc4f305a4fb9c672ca06f290f7c4e261c568f3dea16a410d47/rpds_py-0.30.0-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:922e10f31f303c7c920da8981051ff6d8c1a56207dbdf330d9047f6d30b70e5e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/40/58/442ada3bba6e8e6615fc00483135c14a7538d2ffac30e2d933ccf6852232/rpds_py-0.30.0-cp313-cp313t-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:cdc62c8286ba9bf7f47befdcea13ea0e26bf294bda99758fd90535cbaf408000" },
{ url = "https://mirrors.aliyun.com/pypi/packages/14/14/f59b0127409a33c6ef6f5c1ebd5ad8e32d7861c9c7adfa9a624fc3889f6c/rpds_py-0.30.0-cp313-cp313t-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:47f9a91efc418b54fb8190a6b4aa7813a23fb79c51f4bb84e418f5476c38b8db" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b3/66/e0be3e162ac299b3a22527e8913767d869e6cc75c46bd844aa43fb81ab62/rpds_py-0.30.0-cp313-cp313t-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:1f3587eb9b17f3789ad50824084fa6f81921bbf9a795826570bda82cb3ed91f2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3d/55/fa3b9cf31d0c963ecf1ba777f7cf4b2a2c976795ac430d24a1f43d25a6ba/rpds_py-0.30.0-cp313-cp313t-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:39c02563fc592411c2c61d26b6c5fe1e51eaa44a75aa2c8735ca88b0d9599daa" },
{ url = "https://mirrors.aliyun.com/pypi/packages/60/ca/780cf3b1a32b18c0f05c441958d3758f02544f1d613abf9488cd78876378/rpds_py-0.30.0-cp313-cp313t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:51a1234d8febafdfd33a42d97da7a43f5dcb120c1060e352a3fbc0c6d36e2083" },
{ url = "https://mirrors.aliyun.com/pypi/packages/82/86/d5f2e04f2aa6247c613da0c1dd87fcd08fa17107e858193566048a1e2f0a/rpds_py-0.30.0-cp313-cp313t-manylinux_2_31_riscv64.whl", hash = "sha256:eb2c4071ab598733724c08221091e8d80e89064cd472819285a9ab0f24bcedb9" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4b/9a/453255d2f769fe44e07ea9785c8347edaf867f7026872e76c1ad9f7bed92/rpds_py-0.30.0-cp313-cp313t-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:6bdfdb946967d816e6adf9a3d8201bfad269c67efe6cefd7093ef959683c8de0" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a3/31/622a86cdc0c45d6df0e9ccb6becdba5074735e7033c20e401a6d9d0e2ca0/rpds_py-0.30.0-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:c77afbd5f5250bf27bf516c7c4a016813eb2d3e116139aed0096940c5982da94" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1c/5d/15bbf0fb4a3f58a3b1c67855ec1efcc4ceaef4e86644665fff03e1b66d8d/rpds_py-0.30.0-cp313-cp313t-musllinux_1_2_i686.whl", hash = "sha256:61046904275472a76c8c90c9ccee9013d70a6d0f73eecefd38c1ae7c39045a08" },
{ url = "https://mirrors.aliyun.com/pypi/packages/6d/61/21b8c41f68e60c8cc3b2e25644f0e3681926020f11d06ab0b78e3c6bbff1/rpds_py-0.30.0-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:4c5f36a861bc4b7da6516dbdf302c55313afa09b81931e8280361a4f6c9a2d27" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f9/39/7e067bb06c31de48de3eb200f9fc7c58982a4d3db44b07e73963e10d3be9/rpds_py-0.30.0-cp313-cp313t-win32.whl", hash = "sha256:3d4a69de7a3e50ffc214ae16d79d8fbb0922972da0356dcf4d0fdca2878559c6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0a/4d/222ef0b46443cf4cf46764d9c630f3fe4abaa7245be9417e56e9f52b8f65/rpds_py-0.30.0-cp313-cp313t-win_amd64.whl", hash = "sha256:f14fc5df50a716f7ece6a80b6c78bb35ea2ca47c499e422aa4463455dd96d56d" },
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
]
[[package]]
name = "rsa"
version = "4.9.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
dependencies = [
{ name = "pyasn1" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/da/8a/22b7beea3ee0d44b1916c0c1cb0ee3af23b700b6da9f04991899d0c555d4/rsa-4.9.1.tar.gz", hash = "sha256:e7bdbfdb5497da4c07dfd35530e1a902659db6ff241e39d9953cad06ebd0ae75" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/64/8d/0133e4eb4beed9e425d9a98ed6e081a55d195481b7632472be1af08d2f6b/rsa-4.9.1-py3-none-any.whl", hash = "sha256:68635866661c6836b8d39430f97a996acbd61bfa49406748ea243539fe239762" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
[[package]]
name = "rtfde"
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
version = "0.1.2.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
dependencies = [
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
{ name = "lark" },
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
{ name = "oletools" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/9e/5c/116a016b38af589e8141160bc9b034b73dde2e50c22a921751f4d982a7ca/rtfde-0.1.2.2.tar.gz", hash = "sha256:2f0cd6ecd644071e39452e6fc4f4a1435453af0ec7c90ea86fb4fc96010c7f1b" }
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/14/24/5a653278259be44c1845ddd56dd30cfa7265281ba149b9342b79f9d4f788/rtfde-0.1.2.2-py3-none-any.whl", hash = "sha256:d43868c74f21ae9ea5acbfd4176d5de1f2cfae0ff7f267698471c606287c04ec" },
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
]
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
[[package]]
name = "rtree"
version = "1.4.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/95/09/7302695875a019514de9a5dd17b8320e7a19d6e7bc8f85dcfb79a4ce2da3/rtree-1.4.1.tar.gz", hash = "sha256:c6b1b3550881e57ebe530cc6cffefc87cd9bf49c30b37b894065a9f810875e46" }
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/04/d9/108cd989a4c0954e60b3cdc86fd2826407702b5375f6dfdab2802e5fed98/rtree-1.4.1-py3-none-macosx_10_9_x86_64.whl", hash = "sha256:d672184298527522d4914d8ae53bf76982b86ca420b0acde9298a7a87d81d4a4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f3/cf/2710b6fd6b07ea0aef317b29f335790ba6adf06a28ac236078ed9bd8a91d/rtree-1.4.1-py3-none-macosx_11_0_arm64.whl", hash = "sha256:a7e48d805e12011c2cf739a29d6a60ae852fb1de9fc84220bbcef67e6e595d7d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/55/e1/4d075268a46e68db3cac51846eb6a3ab96ed481c585c5a1ad411b3c23aad/rtree-1.4.1-py3-none-manylinux_2_24_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:efa8c4496e31e9ad58ff6c7df89abceac7022d906cb64a3e18e4fceae6b77f65" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d1/75/e5d44be90525cd28503e7f836d077ae6663ec0687a13ba7810b4114b3668/rtree-1.4.1-py3-none-manylinux_2_24_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:12de4578f1b3381a93a655846900be4e3d5f4cd5e306b8b00aa77c1121dc7e8c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fd/85/b8684f769a142163b52859a38a486493b05bafb4f2fb71d4f945de28ebf9/rtree-1.4.1-py3-none-musllinux_1_2_aarch64.whl", hash = "sha256:b558edda52eca3e6d1ee629042192c65e6b7f2c150d6d6cd207ce82f85be3967" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e9/a4/c2292b95246b9165cc43a0c3757e80995d58bc9b43da5cb47ad6e3535213/rtree-1.4.1-py3-none-musllinux_1_2_x86_64.whl", hash = "sha256:f155bc8d6bac9dcd383481dee8c130947a4866db1d16cb6dff442329a038a0dc" },
{ url = "https://mirrors.aliyun.com/pypi/packages/74/25/5282c8270bfcd620d3e73beb35b40ac4ab00f0a898d98ebeb41ef0989ec8/rtree-1.4.1-py3-none-win_amd64.whl", hash = "sha256:efe125f416fd27150197ab8521158662943a40f87acab8028a1aac4ad667a489" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3f/50/0a9e7e7afe7339bd5e36911f0ceb15fed51945836ed803ae5afd661057fd/rtree-1.4.1-py3-none-win_arm64.whl", hash = "sha256:3d46f55729b28138e897ffef32f7ce93ac335cb67f9120125ad3742a220800f0" },
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "ruamel-base"
version = "1.0.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ea/77/60a0945f4b4eac4b6bd74d1b8e103ae58d0f07b934f962bb4c49e6ec205e/ruamel.base-1.0.0.tar.gz", hash = "sha256:c041333a0f0f00cd6593eb36aa83abb1a9e7544e83ba7a42aa7ac7476cee5cf3" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/1f/e1/5257f5d1636a26fdb50cdcc0e7e5e65d230b88c2dd5090ac797b9e45d1d3/ruamel.base-1.0.0-py3-none-any.whl", hash = "sha256:3613a90afcf0735540804af2a693f630a0bccebefec9b4023a39e88950bb294e" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "ruamel-yaml"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "0.18.17"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
{ name = "ruamel-yaml-clib", marker = "platform_python_implementation == 'CPython'" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/3a/2b/7a1f1ebcd6b3f14febdc003e658778d81e76b40df2267904ee6b13f0c5c6/ruamel_yaml-0.18.17.tar.gz", hash = "sha256:9091cd6e2d93a3a4b157ddb8fabf348c3de7f1fb1381346d985b6b247dcd8d3c" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/af/fe/b6045c782f1fd1ae317d2a6ca1884857ce5c20f59befe6ab25a8603c43a7/ruamel_yaml-0.18.17-py3-none-any.whl", hash = "sha256:9c8ba9eb3e793efdf924b60d521820869d5bf0cb9c6f1b82d82de8295e290b9d" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "ruamel-yaml-clib"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "0.2.15"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ea/97/60fda20e2fb54b83a61ae14648b0817c8f5d84a3821e40bfbdae1437026a/ruamel_yaml_clib-0.2.15.tar.gz", hash = "sha256:46e4cc8c43ef6a94885f72512094e482114a8a706d3c555a34ed4b0d20200600" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/17/5e/2f970ce4c573dc30c2f95825f2691c96d55560268ddc67603dc6ea2dd08e/ruamel_yaml_clib-0.2.15-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:4dcec721fddbb62e60c2801ba08c87010bd6b700054a09998c4d09c08147b8fb" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d6/03/a1baa5b94f71383913f21b96172fb3a2eb5576a4637729adbf7cd9f797f8/ruamel_yaml_clib-0.2.15-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:65f48245279f9bb301d1276f9679b82e4c080a1ae25e679f682ac62446fac471" },
{ url = "https://mirrors.aliyun.com/pypi/packages/dc/19/40d676802390f85784235a05788fd28940923382e3f8b943d25febbb98b7/ruamel_yaml_clib-0.2.15-cp313-cp313-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:46895c17ead5e22bea5e576f1db7e41cb273e8d062c04a6a49013d9f60996c25" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ce/bb/6ef5abfa43b48dd55c30d53e997f8f978722f02add61efba31380d73e42e/ruamel_yaml_clib-0.2.15-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:3eb199178b08956e5be6288ee0b05b2fb0b5c1f309725ad25d9c6ea7e27f962a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ff/5d/e4f84c9c448613e12bd62e90b23aa127ea4c46b697f3d760acc32cb94f25/ruamel_yaml_clib-0.2.15-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:4d1032919280ebc04a80e4fb1e93f7a738129857eaec9448310e638c8bccefcf" },
{ url = "https://mirrors.aliyun.com/pypi/packages/de/4b/e98086e88f76c00c88a6bcf15eae27a1454f661a9eb72b111e6bbb69024d/ruamel_yaml_clib-0.2.15-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:ab0df0648d86a7ecbd9c632e8f8d6b21bb21b5fc9d9e095c796cacf32a728d2d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0c/5c/5964fcd1fd9acc53b7a3a5d9a05ea4f95ead9495d980003a557deb9769c7/ruamel_yaml_clib-0.2.15-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:331fb180858dd8534f0e61aa243b944f25e73a4dae9962bd44c46d1761126bbf" },
{ url = "https://mirrors.aliyun.com/pypi/packages/07/1e/99660f5a30fceb58494598e7d15df883a07292346ef5696f0c0ae5dee8c6/ruamel_yaml_clib-0.2.15-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:fd4c928ddf6bce586285daa6d90680b9c291cfd045fc40aad34e445d57b1bf51" },
{ url = "https://mirrors.aliyun.com/pypi/packages/36/2f/fa0344a9327b58b54970e56a27b32416ffbcfe4dcc0700605516708579b2/ruamel_yaml_clib-0.2.15-cp313-cp313-win32.whl", hash = "sha256:bf0846d629e160223805db9fe8cc7aec16aaa11a07310c50c8c7164efa440aec" },
{ url = "https://mirrors.aliyun.com/pypi/packages/06/c4/c124fbcef0684fcf3c9b72374c2a8c35c94464d8694c50f37eef27f5a145/ruamel_yaml_clib-0.2.15-cp313-cp313-win_amd64.whl", hash = "sha256:45702dfbea1420ba3450bb3dd9a80b33f0badd57539c6aac09f42584303e0db6" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "s3transfer"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "0.16.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "botocore" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/05/04/74127fc843314818edfa81b5540e26dd537353b123a4edc563109d8f17dd/s3transfer-0.16.0.tar.gz", hash = "sha256:8e990f13268025792229cd52fa10cb7163744bf56e719e0b9cb925ab79abf920" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/fc/51/727abb13f44c1fcf6d145979e1535a35794db0f6e450a0cb46aa24732fe2/s3transfer-0.16.0-py3-none-any.whl", hash = "sha256:18e25d66fed509e3868dc1572b3f427ff947dd2c56f844a5bf09481ad3f3b2fe" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "scholarly"
version = "1.7.11"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "arrow" },
{ name = "beautifulsoup4" },
{ name = "bibtexparser" },
{ name = "deprecated" },
{ name = "fake-useragent" },
{ name = "free-proxy" },
{ name = "httpx" },
{ name = "python-dotenv" },
{ name = "requests", extra = ["socks"] },
{ name = "selenium" },
{ name = "sphinx-rtd-theme" },
{ name = "typing-extensions" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/62/2d/38f22adc8abc1166d2c039e477bd8d7782fe32a72f5c80aed94b23348ac1/scholarly-1.7.11.tar.gz", hash = "sha256:2c983dd44d9d9398a6f2605102ae6e5586023b41ebbaec1461917ee48eb153f0" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/b6/23/4340a9068b451b7bb03ff02243bd7aea4c1869781f41e2387c9348629edd/scholarly-1.7.11-py3-none-any.whl", hash = "sha256:be404853e0d020254de32d2050c54dc201f1f36efa4a9d3f8e740d3be4361b20" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "scikit-learn"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "1.8.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "joblib" },
{ name = "numpy" },
Bump infinity to v0.6.11. Requires python>=3.11 (#11814) ### What problem does this PR solve? Bump infinity to v0.6.11. Requires python>=3.11 ### Type of change - [x] Refactoring
2025-12-09 16:23:37 +08:00
{ name = "scipy" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "threadpoolctl" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/0e/d4/40988bf3b8e34feec1d0e6a051446b1f66225f8529b9309becaeef62b6c4/scikit_learn-1.8.0.tar.gz", hash = "sha256:9bccbb3b40e3de10351f8f5068e105d0f4083b1a65fa07b6634fbc401a6287fd" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/03/aa/e22e0768512ce9255eba34775be2e85c2048da73da1193e841707f8f039c/scikit_learn-1.8.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:0d6ae97234d5d7079dc0040990a6f7aeb97cb7fa7e8945f1999a429b23569e0a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/58/37/31b83b2594105f61a381fc74ca19e8780ee923be2d496fcd8d2e1147bd99/scikit_learn-1.8.0-cp313-cp313-macosx_12_0_arm64.whl", hash = "sha256:edec98c5e7c128328124a029bceb09eda2d526997780fef8d65e9a69eead963e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2d/5a/3f1caed8765f33eabb723596666da4ebbf43d11e96550fb18bdec42b467b/scikit_learn-1.8.0-cp313-cp313-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:74b66d8689d52ed04c271e1329f0c61635bcaf5b926db9b12d58914cdc01fe57" },
{ url = "https://mirrors.aliyun.com/pypi/packages/38/cf/06896db3f71c75902a8e9943b444a56e727418f6b4b4a90c98c934f51ed4/scikit_learn-1.8.0-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:8fdf95767f989b0cfedb85f7ed8ca215d4be728031f56ff5a519ee1e3276dc2e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1c/f9/9b7563caf3ec8873e17a31401858efab6b39a882daf6c1bfa88879c0aa11/scikit_learn-1.8.0-cp313-cp313-win_amd64.whl", hash = "sha256:2de443b9373b3b615aec1bb57f9baa6bb3a9bd093f1269ba95c17d870422b271" },
{ url = "https://mirrors.aliyun.com/pypi/packages/49/bd/1f4001503650e72c4f6009ac0c4413cb17d2d601cef6f71c0453da2732fc/scikit_learn-1.8.0-cp313-cp313-win_arm64.whl", hash = "sha256:eddde82a035681427cbedded4e6eff5e57fa59216c2e3e90b10b19ab1d0a65c3" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d2/7d/a630359fc9dcc95496588c8d8e3245cc8fd81980251079bc09c70d41d951/scikit_learn-1.8.0-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:7cc267b6108f0a1499a734167282c00c4ebf61328566b55ef262d48e9849c735" },
{ url = "https://mirrors.aliyun.com/pypi/packages/cc/56/a0c86f6930cfcd1c7054a2bc417e26960bb88d32444fe7f71d5c2cfae891/scikit_learn-1.8.0-cp313-cp313t-macosx_12_0_arm64.whl", hash = "sha256:fe1c011a640a9f0791146011dfd3c7d9669785f9fed2b2a5f9e207536cf5c2fd" },
{ url = "https://mirrors.aliyun.com/pypi/packages/46/1e/05962ea1cebc1cf3876667ecb14c283ef755bf409993c5946ade3b77e303/scikit_learn-1.8.0-cp313-cp313t-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:72358cce49465d140cc4e7792015bb1f0296a9742d5622c67e31399b75468b9e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fe/56/a85473cd75f200c9759e3a5f0bcab2d116c92a8a02ee08ccd73b870f8bb4/scikit_learn-1.8.0-cp313-cp313t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:80832434a6cc114f5219211eec13dcbc16c2bac0e31ef64c6d346cde3cf054cb" },
{ url = "https://mirrors.aliyun.com/pypi/packages/cc/b7/64d8cfa896c64435ae57f4917a548d7ac7a44762ff9802f75a79b77cb633/scikit_learn-1.8.0-cp313-cp313t-win_amd64.whl", hash = "sha256:ee787491dbfe082d9c3013f01f5991658b0f38aa8177e4cd4bf434c58f551702" },
{ url = "https://mirrors.aliyun.com/pypi/packages/5e/37/e192ea709551799379958b4c4771ec507347027bb7c942662c7fbeba31cb/scikit_learn-1.8.0-cp313-cp313t-win_arm64.whl", hash = "sha256:bf97c10a3f5a7543f9b88cbf488d33d175e9146115a451ae34568597ba33dcde" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
[[package]]
name = "scipy"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.17.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
dependencies = [
Bump infinity to v0.6.11. Requires python>=3.11 (#11814) ### What problem does this PR solve? Bump infinity to v0.6.11. Requires python>=3.11 ### Type of change - [x] Refactoring
2025-12-09 16:23:37 +08:00
{ name = "numpy" },
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/56/3e/9cca699f3486ce6bc12ff46dc2031f1ec8eb9ccc9a320fdaf925f1417426/scipy-1.17.0.tar.gz", hash = "sha256:2591060c8e648d8b96439e111ac41fd8342fdeff1876be2e19dea3fe8930454e" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/0c/51/3468fdfd49387ddefee1636f5cf6d03ce603b75205bf439bbf0e62069bfd/scipy-1.17.0-cp313-cp313-macosx_10_14_x86_64.whl", hash = "sha256:65ec32f3d32dfc48c72df4291345dae4f048749bc8d5203ee0a3f347f96c5ce6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b2/9a/9406aec58268d437636069419e6977af953d1e246df941d42d3720b7277b/scipy-1.17.0-cp313-cp313-macosx_12_0_arm64.whl", hash = "sha256:1f9586a58039d7229ce77b52f8472c972448cded5736eaf102d5658bbac4c269" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4f/98/e7342709e17afdfd1b26b56ae499ef4939b45a23a00e471dfb5375eea205/scipy-1.17.0-cp313-cp313-macosx_14_0_arm64.whl", hash = "sha256:9fad7d3578c877d606b1150135c2639e9de9cecd3705caa37b66862977cc3e72" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fd/0e/9eeeb5357a64fd157cbe0302c213517c541cc16b8486d82de251f3c68ede/scipy-1.17.0-cp313-cp313-macosx_14_0_x86_64.whl", hash = "sha256:423ca1f6584fc03936972b5f7c06961670dbba9f234e71676a7c7ccf938a0d61" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c9/10/be13397a0e434f98e0c79552b2b584ae5bb1c8b2be95db421533bbca5369/scipy-1.17.0-cp313-cp313-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:fe508b5690e9eaaa9467fc047f833af58f1152ae51a0d0aed67aa5801f4dd7d6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/63/1e/12fbf2a3bb240161651c94bb5cdd0eae5d4e8cc6eaeceb74ab07b12a753d/scipy-1.17.0-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:6680f2dfd4f6182e7d6db161344537da644d1cf85cf293f015c60a17ecf08752" },
{ url = "https://mirrors.aliyun.com/pypi/packages/19/5b/1a63923e23ccd20bd32156d7dd708af5bbde410daa993aa2500c847ab2d2/scipy-1.17.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:eec3842ec9ac9de5917899b277428886042a93db0b227ebbe3a333b64ec7643d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/39/22/b5da95d74edcf81e540e467202a988c50fef41bd2011f46e05f72ba07df6/scipy-1.17.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:d7425fcafbc09a03731e1bc05581f5fad988e48c6a861f441b7ab729a49a55ea" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b9/b6/8ac583d6da79e7b9e520579f03007cb006f063642afd6b2eeb16b890bf93/scipy-1.17.0-cp313-cp313-win_amd64.whl", hash = "sha256:87b411e42b425b84777718cc41516b8a7e0795abfa8e8e1d573bf0ef014f0812" },
{ url = "https://mirrors.aliyun.com/pypi/packages/55/fb/7db19e0b3e52f882b420417644ec81dd57eeef1bd1705b6f689d8ff93541/scipy-1.17.0-cp313-cp313-win_arm64.whl", hash = "sha256:357ca001c6e37601066092e7c89cca2f1ce74e2a520ca78d063a6d2201101df2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/20/b6/7feaa252c21cc7aff335c6c55e1b90ab3e3306da3f048109b8b639b94648/scipy-1.17.0-cp313-cp313t-macosx_10_14_x86_64.whl", hash = "sha256:ec0827aa4d36cb79ff1b81de898e948a51ac0b9b1c43e4a372c0508c38c0f9a3" },
{ url = "https://mirrors.aliyun.com/pypi/packages/76/bb/bbb392005abce039fb7e672cb78ac7d158700e826b0515cab6b5b60c26fb/scipy-1.17.0-cp313-cp313t-macosx_12_0_arm64.whl", hash = "sha256:819fc26862b4b3c73a60d486dbb919202f3d6d98c87cf20c223511429f2d1a97" },
{ url = "https://mirrors.aliyun.com/pypi/packages/37/da/9d33196ecc99fba16a409c691ed464a3a283ac454a34a13a3a57c0d66f3a/scipy-1.17.0-cp313-cp313t-macosx_14_0_arm64.whl", hash = "sha256:363ad4ae2853d88ebcde3ae6ec46ccca903ea9835ee8ba543f12f575e7b07e4e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/56/9d/f4b184f6ddb28e9a5caea36a6f98e8ecd2a524f9127354087ce780885d83/scipy-1.17.0-cp313-cp313t-macosx_14_0_x86_64.whl", hash = "sha256:979c3a0ff8e5ba254d45d59ebd38cde48fce4f10b5125c680c7a4bfe177aab07" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9b/9d/025cccdd738a72140efc582b1641d0dd4caf2e86c3fb127568dc80444e6e/scipy-1.17.0-cp313-cp313t-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:130d12926ae34399d157de777472bf82e9061c60cc081372b3118edacafe1d00" },
{ url = "https://mirrors.aliyun.com/pypi/packages/48/5f/09b879619f8bca15ce392bfc1894bd9c54377e01d1b3f2f3b595a1b4d945/scipy-1.17.0-cp313-cp313t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:6e886000eb4919eae3a44f035e63f0fd8b651234117e8f6f29bad1cd26e7bc45" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f2/9a/f0f0a9f0aa079d2f106555b984ff0fbb11a837df280f04f71f056ea9c6e4/scipy-1.17.0-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:13c4096ac6bc31d706018f06a49abe0485f96499deb82066b94d19b02f664209" },
{ url = "https://mirrors.aliyun.com/pypi/packages/90/b8/4f0f5cf0c5ea4d7548424e6533e6b17d164f34a6e2fb2e43ffebb6697b06/scipy-1.17.0-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:cacbaddd91fcffde703934897c5cd2c7cb0371fac195d383f4e1f1c5d3f3bd04" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f9/cc/2bd59140ed3b2fa2882fb15da0a9cb1b5a6443d67cfd0d98d4cec83a57ec/scipy-1.17.0-cp313-cp313t-win_amd64.whl", hash = "sha256:edce1a1cf66298cccdc48a1bdf8fb10a3bf58e8b58d6c3883dd1530e103f87c0" },
{ url = "https://mirrors.aliyun.com/pypi/packages/13/1b/c87cc44a0d2c7aaf0f003aef2904c3d097b422a96c7e7c07f5efd9073c1b/scipy-1.17.0-cp313-cp313t-win_arm64.whl", hash = "sha256:30509da9dbec1c2ed8f168b8d8aa853bc6723fede1dbc23c7d43a56f5ab72a67" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
[[package]]
name = "screeninfo"
version = "0.8.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
dependencies = [
{ name = "cython", marker = "sys_platform == 'darwin'" },
{ name = "pyobjc-framework-cocoa", marker = "sys_platform == 'darwin'" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ec/bb/e69e5e628d43f118e0af4fc063c20058faa8635c95a1296764acc8167e27/screeninfo-0.8.1.tar.gz", hash = "sha256:9983076bcc7e34402a1a9e4d7dabf3729411fd2abb3f3b4be7eba73519cd2ed1" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/6e/bf/c5205d480307bef660e56544b9e3d7ff687da776abb30c9cb3f330887570/screeninfo-0.8.1-py3-none-any.whl", hash = "sha256:e97d6b173856edcfa3bd282f81deb528188aff14b11ec3e195584e7641be733c" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "seaborn"
version = "0.13.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "matplotlib" },
{ name = "numpy" },
{ name = "pandas" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/86/59/a451d7420a77ab0b98f7affa3a1d78a313d2f7281a57afb1a34bae8ab412/seaborn-0.13.2.tar.gz", hash = "sha256:93e60a40988f4d65e9f4885df477e2fdaff6b73a9ded434c1ab356dd57eefff7" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/83/11/00d3c3dfc25ad54e731d91449895a79e4bf2384dc3ac01809010ba88f6d5/seaborn-0.13.2-py3-none-any.whl", hash = "sha256:636f8336facf092165e27924f223d3c62ca560b1f2bb5dff7ab7fad265361987" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "selenium"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "4.32.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "certifi" },
{ name = "trio" },
{ name = "trio-websocket" },
{ name = "typing-extensions" },
{ name = "urllib3", extra = ["socks"] },
{ name = "websocket-client" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/54/2d/fafffe946099033ccf22bf89e12eede14c1d3c5936110c5f6f2b9830722c/selenium-4.32.0.tar.gz", hash = "sha256:b9509bef4056f4083772abb1ae19ff57247d617a29255384b26be6956615b206" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ea/37/d07ed9d13e571b2115d4ed6956d156c66816ceec0b03b2e463e80d09f572/selenium-4.32.0-py3-none-any.whl", hash = "sha256:c4d9613f8a45693d61530c9660560fadb52db7d730237bc788ddedf442391f97" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "selenium-wire"
version = "5.1.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "blinker" },
{ name = "brotli" },
{ name = "certifi" },
{ name = "h2" },
{ name = "hyperframe" },
{ name = "kaitaistruct" },
{ name = "pyasn1" },
{ name = "pydivert", marker = "sys_platform == 'win32'" },
{ name = "pyopenssl" },
{ name = "pyparsing" },
{ name = "pysocks" },
{ name = "selenium" },
{ name = "wsproto" },
{ name = "zstandard" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/9f/00/60b39e8a1efe6919d1390f07d84a3eeba4aeae5b829f2f848344c798f783/selenium-wire-5.1.0.tar.gz", hash = "sha256:b1cd4eae44d9959381abe3bb186472520d063c658e279f98555def3d4e6dd29b" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/eb/7e/9548b365eab16730b6a8da25c6e1f83f3b84fb6092ecd2d4d69933d08a45/selenium_wire-5.1.0-py3-none-any.whl", hash = "sha256:fbf930d9992f8b6d24bb16b3e6221bab596a41f6ae7548270b7d5a92f3402622" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "setuptools"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "80.9.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/18/5d/3bf57dcd21979b887f014ea83c24ae194cfcd12b9e0fda66b957c69d1fca/setuptools-80.9.0.tar.gz", hash = "sha256:f36b47402ecde768dbfafc46e8e4207b4360c654f1f3bb84475f0a28628fb19c" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/a3/dc/17031897dae0efacfea57dfd3a82fdd2a2aeb58e0ff71b77b87e44edc772/setuptools-80.9.0-py3-none-any.whl", hash = "sha256:062d34222ad13e0cc312a4c02d73f059e86a4acbfbdea8f8f76b28c99f306922" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "sgmllib3k"
version = "1.0.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/9e/bd/3704a8c3e0942d711c1299ebf7b9091930adae6675d7c8f476a7ce48653c/sgmllib3k-1.0.0.tar.gz", hash = "sha256:7868fb1c8bfa764c1ac563d3cf369c381d1325d36124933a726f29fcdaa812e9" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "shapely"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "2.1.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "numpy" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/4d/bc/0989043118a27cccb4e906a46b7565ce36ca7b57f5a18b78f4f1b0f72d9d/shapely-2.1.2.tar.gz", hash = "sha256:2ed4ecb28320a433db18a5bf029986aa8afcfd740745e78847e330d5d94922a9" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/c3/90/98ef257c23c46425dc4d1d31005ad7c8d649fe423a38b917db02c30f1f5a/shapely-2.1.2-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:b510dda1a3672d6879beb319bc7c5fd302c6c354584690973c838f46ec3e0fa8" },
{ url = "https://mirrors.aliyun.com/pypi/packages/6d/ab/0bee5a830d209adcd3a01f2d4b70e587cdd9fd7380d5198c064091005af8/shapely-2.1.2-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:8cff473e81017594d20ec55d86b54bc635544897e13a7cfc12e36909c5309a2a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2d/5e/7d7f54ba960c13302584c73704d8c4d15404a51024631adb60b126a4ae88/shapely-2.1.2-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:fe7b77dc63d707c09726b7908f575fc04ff1d1ad0f3fb92aec212396bc6cfe5e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f2/a2/83fc37e2a58090e3d2ff79175a95493c664bcd0b653dd75cb9134645a4e5/shapely-2.1.2-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:7ed1a5bbfb386ee8332713bf7508bc24e32d24b74fc9a7b9f8529a55db9f4ee6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/44/2b/578faf235a5b09f16b5f02833c53822294d7f21b242f8e2d0cf03fb64321/shapely-2.1.2-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:a84e0582858d841d54355246ddfcbd1fce3179f185da7470f41ce39d001ee1af" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4d/04/167f096386120f692cc4ca02f75a17b961858997a95e67a3cb6a7bbd6b53/shapely-2.1.2-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:dc3487447a43d42adcdf52d7ac73804f2312cbfa5d433a7d2c506dcab0033dfd" },
{ url = "https://mirrors.aliyun.com/pypi/packages/48/74/fb402c5a6235d1c65a97348b48cdedb75fb19eca2b1d66d04969fc1c6091/shapely-2.1.2-cp313-cp313-win32.whl", hash = "sha256:9c3a3c648aedc9f99c09263b39f2d8252f199cb3ac154fadc173283d7d111350" },
{ url = "https://mirrors.aliyun.com/pypi/packages/41/47/3647fe7ad990af60ad98b889657a976042c9988c2807cf322a9d6685f462/shapely-2.1.2-cp313-cp313-win_amd64.whl", hash = "sha256:ca2591bff6645c216695bdf1614fca9c82ea1144d4a7591a466fef64f28f0715" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3c/49/63953754faa51ffe7d8189bfbe9ca34def29f8c0e34c67cbe2a2795f269d/shapely-2.1.2-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:2d93d23bdd2ed9dc157b46bc2f19b7da143ca8714464249bef6771c679d5ff40" },
{ url = "https://mirrors.aliyun.com/pypi/packages/7f/ee/dce001c1984052970ff60eb4727164892fb2d08052c575042a47f5a9e88f/shapely-2.1.2-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:01d0d304b25634d60bd7cf291828119ab55a3bab87dc4af1e44b07fb225f188b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/da/e7/fc4e9a19929522877fa602f705706b96e78376afb7fad09cad5b9af1553c/shapely-2.1.2-cp313-cp313t-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:8d8382dd120d64b03698b7298b89611a6ea6f55ada9d39942838b79c9bc89801" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a1/18/7519a25db21847b525696883ddc8e6a0ecaa36159ea88e0fef11466384d0/shapely-2.1.2-cp313-cp313t-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:19efa3611eef966e776183e338b2d7ea43569ae99ab34f8d17c2c054d3205cc0" },
{ url = "https://mirrors.aliyun.com/pypi/packages/48/de/b59a620b1f3a129c3fecc2737104a0a7e04e79335bd3b0a1f1609744cf17/shapely-2.1.2-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:346ec0c1a0fcd32f57f00e4134d1200e14bf3f5ae12af87ba83ca275c502498c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/96/b3/c6655ee7232b417562bae192ae0d3ceaadb1cc0ffc2088a2ddf415456cc2/shapely-2.1.2-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:6305993a35989391bd3476ee538a5c9a845861462327efe00dd11a5c8c709a99" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a0/8e/605c76808d73503c9333af8f6cbe7e1354d2d238bda5f88eea36bfe0f42a/shapely-2.1.2-cp313-cp313t-win32.whl", hash = "sha256:c8876673449f3401f278c86eb33224c5764582f72b653a415d0e6672fde887bf" },
{ url = "https://mirrors.aliyun.com/pypi/packages/36/f7/d317eb232352a1f1444d11002d477e54514a4a6045536d49d0c59783c0da/shapely-2.1.2-cp313-cp313t-win_amd64.whl", hash = "sha256:4a44bc62a10d84c11a7a3d7c1c4fe857f7477c3506e24c9062da0db0ae0c449c" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "shellingham"
version = "1.5.4"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/58/15/8b3609fd3830ef7b27b655beb4b4e9c62313a4e8da8c676e142cc210d58e/shellingham-1.5.4.tar.gz", hash = "sha256:8dbca0739d487e5bd35ab3ca4b36e11c4078f3a234bfce294b0a0291363404de" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/e0/f9/0595336914c5619e5f28a1fb793285925a8cd4b432c9da0a987836c7f822/shellingham-1.5.4-py2.py3-none-any.whl", hash = "sha256:7ecfff8f2fd72616f7481040475a65b2bf8af90a56c89140852d1120324e8686" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "six"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "1.17.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/94/e7/b2c673351809dca68a0e064b6af791aa332cf192da575fd474ed7d6f16a2/six-1.17.0.tar.gz", hash = "sha256:ff70335d468e7eb6ec65b95b99d3a2836546063f63acc5171de367e834932a81" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/b7/ce/149a00dd41f10bc29e5921b496af8b574d8413afcd5e30dfa0ed46c2cc5e/six-1.17.0-py2.py3-none-any.whl", hash = "sha256:4721f391ed90541fddacab5acf947aa0d3dc7d27b2e1e8eda2be8970586c3274" },
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
]
[[package]]
name = "slack-sdk"
version = "3.37.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/8e/c2/0a174a155623d7dc3ed4d1360cdf755590acdc2c3fc9ce0d2340f468909f/slack_sdk-3.37.0.tar.gz", hash = "sha256:242d6cffbd9e843af807487ff04853189b812081aeaa22f90a8f159f20220ed9" }
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/07/fd/a502ee24d8c7d12a8f749878ae0949b8eeb50aeac22dc5a613d417a256d0/slack_sdk-3.37.0-py2.py3-none-any.whl", hash = "sha256:e108a0836eafda74d8a95e76c12c2bcb010e645d504d8497451e4c7ebb229c87" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "smart-open"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "7.5.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "wrapt" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/67/9a/0a7acb748b86e2922982366d780ca4b16c33f7246fa5860d26005c97e4f3/smart_open-7.5.0.tar.gz", hash = "sha256:f394b143851d8091011832ac8113ea4aba6b92e6c35f6e677ddaaccb169d7cb9" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ad/95/bc978be7ea0babf2fb48a414b6afaad414c6a9e8b1eafc5b8a53c030381a/smart_open-7.5.0-py3-none-any.whl", hash = "sha256:87e695c5148bbb988f15cec00971602765874163be85acb1c9fb8abc012e6599" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "sniffio"
version = "1.3.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a2/87/a6771e1546d97e7e041b6ae58d80074f81b7d5121207425c964ddf5cfdbd/sniffio-1.3.1.tar.gz", hash = "sha256:f4324edc670a0f49750a81b895f35c3adb843cca46f0530f79fc1babb23789dc" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/e9/44/75a9c9421471a6c4805dbf2356f7c181a29c1879239abab1ea2cc8f38b40/sniffio-1.3.1-py3-none-any.whl", hash = "sha256:2f6da418d1f1e0fddd844478f41680e794e6051915791a034ff65e5f100525a2" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "snowballstemmer"
version = "2.2.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/44/7b/af302bebf22c749c56c9c3e8ae13190b5b5db37a33d9068652e8f73b7089/snowballstemmer-2.2.0.tar.gz", hash = "sha256:09b16deb8547d3412ad7b590689584cd0fe25ec8db3be37788be3810cbf19cb1" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ed/dc/c02e01294f7265e63a7315fe086dd1df7dacb9f840a804da846b96d01b96/snowballstemmer-2.2.0-py2.py3-none-any.whl", hash = "sha256:c8e1716e83cc398ae16824e5572ae04e0d9fc2c6b985fb0f900f5f0c96ecba1a" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Env: Update dependencies for proxy support (#9519) ### What problem does this PR solve? - Update httpx dependency to include socks support in pyproject.toml - Update lockfile with new socksio dependency ### Type of change - [x] Update dependencies for proxy support
2025-08-18 12:04:16 +08:00
[[package]]
name = "socksio"
version = "1.0.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f8/5c/48a7d9495be3d1c651198fd99dbb6ce190e2274d0f28b9051307bdec6b85/socksio-1.0.0.tar.gz", hash = "sha256:f88beb3da5b5c38b9890469de67d0cb0f9d494b78b106ca1845f96c10b91c4ac" }
Env: Update dependencies for proxy support (#9519) ### What problem does this PR solve? - Update httpx dependency to include socks support in pyproject.toml - Update lockfile with new socksio dependency ### Type of change - [x] Update dependencies for proxy support
2025-08-18 12:04:16 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/37/c3/6eeb6034408dac0fa653d126c9204ade96b819c936e136c5e8a6897eee9c/socksio-1.0.0-py3-none-any.whl", hash = "sha256:95dc1f15f9b34e8d7b16f06d74b8ccf48f609af32ab33c608d08761c5dcbb1f3" },
Env: Update dependencies for proxy support (#9519) ### What problem does this PR solve? - Update httpx dependency to include socks support in pyproject.toml - Update lockfile with new socksio dependency ### Type of change - [x] Update dependencies for proxy support
2025-08-18 12:04:16 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "sortedcontainers"
version = "2.4.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/e8/c4/ba2f8066cceb6f23394729afe52f3bf7adec04bf9ed2c820b39e19299111/sortedcontainers-2.4.0.tar.gz", hash = "sha256:25caa5a06cc30b6b83d11423433f65d1f9d76c4c6a0c90e3379eaa43b9bfdb88" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/32/46/9cb0e58b2deb7f82b84065f37f3bffeb12413f947f9388e4cac22c4621ce/sortedcontainers-2.4.0-py2.py3-none-any.whl", hash = "sha256:a163dcaede0f1c021485e957a39245190e74249897e2ae4b2aa38595db237ee0" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "soupsieve"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "2.8.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/89/23/adf3796d740536d63a6fbda113d07e60c734b6ed5d3058d1e47fc0495e47/soupsieve-2.8.1.tar.gz", hash = "sha256:4cf733bc50fa805f5df4b8ef4740fc0e0fa6218cf3006269afd3f9d6d80fd350" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/48/f3/b67d6ea49ca9154453b6d70b34ea22f3996b9fa55da105a79d8732227adc/soupsieve-2.8.1-py3-none-any.whl", hash = "sha256:a11fe2a6f3d76ab3cf2de04eb339c1be5b506a8a47f2ceb6d139803177f85434" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
[[package]]
name = "spacy"
version = "3.8.14"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
dependencies = [
{ name = "catalogue" },
{ name = "confection" },
{ name = "cymem" },
{ name = "jinja2" },
{ name = "murmurhash" },
{ name = "numpy" },
{ name = "packaging" },
{ name = "preshed" },
{ name = "pydantic" },
{ name = "requests" },
{ name = "setuptools" },
{ name = "spacy-legacy" },
{ name = "spacy-loggers" },
{ name = "srsly" },
{ name = "thinc" },
{ name = "tqdm" },
{ name = "typer" },
{ name = "wasabi" },
{ name = "weasel" },
]
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/1b/e5/822bbdfa459fee863ef2e9879a34b0ae5db7cd1e3eb76d32c766f19222e9/spacy-3.8.14-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:2b4f60fa8b9641a5e93e7a96db0cdd106d05d61756bf1d0ddcd1705ad347909a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/7e/de/0e512154113e1f341567f2b9341835775e4180c180221e60faedaebb2f65/spacy-3.8.14-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:0860c57220c633ccb20468bcd64bfb0d28908990c371a8857951d093a148dc8e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0c/4f/29c7e56afc7db07348a9e0efe0243b5eef465d5dc3d56433f164378c3fa6/spacy-3.8.14-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:c24620b7dba879c69cebc51ef3b1107d4d4e44a1e0d4baa439372887d00c3fd9" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1e/ce/cae678f664d5467016819253f5d6e52f8e68a12d8e799b651d73ec2a9a4b/spacy-3.8.14-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:9699c1248d115d5825987c287a6f6acd66386ef3ebee7994ee67ba093e932c59" },
{ url = "https://mirrors.aliyun.com/pypi/packages/04/d4/419868afd449bdd367df005932537eea66c71e97c899ba278f3124933f3c/spacy-3.8.14-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:042d799e342fdb6bb5b02a4213a95acc9116c40ed3c849bb0a8296fbe648ec22" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ec/53/df5c1fee45f200b749ba72eeb536fbb2c545fc56230324954263b2f3be00/spacy-3.8.14-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:69b2264294097336e86832e8663f1ab3a7215621184863c96c082ab17ee11937" },
{ url = "https://mirrors.aliyun.com/pypi/packages/12/c2/f1882ec2f5cc9c4e73cf2132997a03c397d7ceeb5ee7f7bb878b51a16365/spacy-3.8.14-cp313-cp313-win_amd64.whl", hash = "sha256:4b6d4f20e291a7c70e37de2f246622b44a0ce82efaa710c9801c6bd599e75177" },
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
]
[[package]]
name = "spacy-legacy"
version = "3.0.12"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/d9/79/91f9d7cc8db5642acad830dcc4b49ba65a7790152832c4eceb305e46d681/spacy-legacy-3.0.12.tar.gz", hash = "sha256:b37d6e0c9b6e1d7ca1cf5bc7152ab64a4c4671f59c85adaf7a3fcb870357a774" }
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/c3/55/12e842c70ff8828e34e543a2c7176dac4da006ca6901c9e8b43efab8bc6b/spacy_legacy-3.0.12-py2.py3-none-any.whl", hash = "sha256:476e3bd0d05f8c339ed60f40986c07387c0a71479245d6d0f4298dbd52cda55f" },
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
]
[[package]]
name = "spacy-loggers"
version = "1.0.5"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/67/3d/926db774c9c98acf66cb4ed7faf6c377746f3e00b84b700d0868b95d0712/spacy-loggers-1.0.5.tar.gz", hash = "sha256:d60b0bdbf915a60e516cc2e653baeff946f0cfc461b452d11a4d5458c6fe5f24" }
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/33/78/d1a1a026ef3af911159398c939b1509d5c36fe524c7b644f34a5146c4e16/spacy_loggers-1.0.5-py3-none-any.whl", hash = "sha256:196284c9c446cc0cdb944005384270d775fdeaf4f494d8e269466cfa497ef645" },
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
]
Add tavily as web searh tool. (#5349) ### What problem does this PR solve? #5198 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-02-26 10:21:04 +08:00
[[package]]
name = "sphinx"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "8.2.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Bump infinity to v0.6.11. Requires python>=3.11 (#11814) ### What problem does this PR solve? Bump infinity to v0.6.11. Requires python>=3.11 ### Type of change - [x] Refactoring
2025-12-09 16:23:37 +08:00
dependencies = [
{ name = "alabaster" },
{ name = "babel" },
{ name = "colorama", marker = "sys_platform == 'win32'" },
{ name = "docutils" },
{ name = "imagesize" },
{ name = "jinja2" },
{ name = "packaging" },
{ name = "pygments" },
{ name = "requests" },
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
{ name = "roman-numerals-py" },
Bump infinity to v0.6.11. Requires python>=3.11 (#11814) ### What problem does this PR solve? Bump infinity to v0.6.11. Requires python>=3.11 ### Type of change - [x] Refactoring
2025-12-09 16:23:37 +08:00
{ name = "snowballstemmer" },
{ name = "sphinxcontrib-applehelp" },
{ name = "sphinxcontrib-devhelp" },
{ name = "sphinxcontrib-htmlhelp" },
{ name = "sphinxcontrib-jsmath" },
{ name = "sphinxcontrib-qthelp" },
{ name = "sphinxcontrib-serializinghtml" },
Add tavily as web searh tool. (#5349) ### What problem does this PR solve? #5198 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-02-26 10:21:04 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/38/ad/4360e50ed56cb483667b8e6dadf2d3fda62359593faabbe749a27c4eaca6/sphinx-8.2.3.tar.gz", hash = "sha256:398ad29dee7f63a75888314e9424d40f52ce5a6a87ae88e7071e80af296ec348" }
Add tavily as web searh tool. (#5349) ### What problem does this PR solve? #5198 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-02-26 10:21:04 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/31/53/136e9eca6e0b9dc0e1962e2c908fbea2e5ac000c2a2fbd9a35797958c48b/sphinx-8.2.3-py3-none-any.whl", hash = "sha256:4405915165f13521d875a8c29c8970800a0141c14cc5416a38feca4ea5d9b9c3" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "sphinx-rtd-theme"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "3.0.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "docutils" },
Bump infinity to v0.6.11. Requires python>=3.11 (#11814) ### What problem does this PR solve? Bump infinity to v0.6.11. Requires python>=3.11 ### Type of change - [x] Refactoring
2025-12-09 16:23:37 +08:00
{ name = "sphinx" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "sphinxcontrib-jquery" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/91/44/c97faec644d29a5ceddd3020ae2edffa69e7d00054a8c7a6021e82f20335/sphinx_rtd_theme-3.0.2.tar.gz", hash = "sha256:b7457bc25dda723b20b086a670b9953c859eab60a2a03ee8eb2bb23e176e5f85" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/85/77/46e3bac77b82b4df5bb5b61f2de98637724f246b4966cfc34bc5895d852a/sphinx_rtd_theme-3.0.2-py2.py3-none-any.whl", hash = "sha256:422ccc750c3a3a311de4ae327e82affdaf59eb695ba4936538552f3b00f4ee13" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "sphinxcontrib-applehelp"
version = "2.0.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ba/6e/b837e84a1a704953c62ef8776d45c3e8d759876b4a84fe14eba2859106fe/sphinxcontrib_applehelp-2.0.0.tar.gz", hash = "sha256:2f29ef331735ce958efa4734873f084941970894c6090408b079c61b2e1c06d1" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/5d/85/9ebeae2f76e9e77b952f4b274c27238156eae7979c5421fba91a28f4970d/sphinxcontrib_applehelp-2.0.0-py3-none-any.whl", hash = "sha256:4cd3f0ec4ac5dd9c17ec65e9ab272c9b867ea77425228e68ecf08d6b28ddbdb5" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "sphinxcontrib-devhelp"
version = "2.0.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/f6/d2/5beee64d3e4e747f316bae86b55943f51e82bb86ecd325883ef65741e7da/sphinxcontrib_devhelp-2.0.0.tar.gz", hash = "sha256:411f5d96d445d1d73bb5d52133377b4248ec79db5c793ce7dbe59e074b4dd1ad" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/35/7a/987e583882f985fe4d7323774889ec58049171828b58c2217e7f79cdf44e/sphinxcontrib_devhelp-2.0.0-py3-none-any.whl", hash = "sha256:aefb8b83854e4b0998877524d1029fd3e6879210422ee3780459e28a1f03a8a2" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "sphinxcontrib-htmlhelp"
version = "2.1.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/43/93/983afd9aa001e5201eab16b5a444ed5b9b0a7a010541e0ddfbbfd0b2470c/sphinxcontrib_htmlhelp-2.1.0.tar.gz", hash = "sha256:c9e2916ace8aad64cc13a0d233ee22317f2b9025b9cf3295249fa985cc7082e9" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/0a/7b/18a8c0bcec9182c05a0b3ec2a776bba4ead82750a55ff798e8d406dae604/sphinxcontrib_htmlhelp-2.1.0-py3-none-any.whl", hash = "sha256:166759820b47002d22914d64a075ce08f4c46818e17cfc9470a9786b759b19f8" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "sphinxcontrib-jquery"
version = "4.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
Bump infinity to v0.6.11. Requires python>=3.11 (#11814) ### What problem does this PR solve? Bump infinity to v0.6.11. Requires python>=3.11 ### Type of change - [x] Refactoring
2025-12-09 16:23:37 +08:00
{ name = "sphinx" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/de/f3/aa67467e051df70a6330fe7770894b3e4f09436dea6881ae0b4f3d87cad8/sphinxcontrib-jquery-4.1.tar.gz", hash = "sha256:1620739f04e36a2c779f1a131a2dfd49b2fd07351bf1968ced074365933abc7a" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/76/85/749bd22d1a68db7291c89e2ebca53f4306c3f205853cf31e9de279034c3c/sphinxcontrib_jquery-4.1-py2.py3-none-any.whl", hash = "sha256:f936030d7d0147dd026a4f2b5a57343d233f1fc7b363f68b3d4f1cb0993878ae" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "sphinxcontrib-jsmath"
version = "1.0.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/b2/e8/9ed3830aeed71f17c026a07a5097edcf44b692850ef215b161b8ad875729/sphinxcontrib-jsmath-1.0.1.tar.gz", hash = "sha256:a9925e4a4587247ed2191a22df5f6970656cb8ca2bd6284309578f2153e0c4b8" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/c2/42/4c8646762ee83602e3fb3fbe774c2fac12f317deb0b5dbeeedd2d3ba4b77/sphinxcontrib_jsmath-1.0.1-py2.py3-none-any.whl", hash = "sha256:2ec2eaebfb78f3f2078e73666b1415417a116cc848b72e5172e596c871103178" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "sphinxcontrib-qthelp"
version = "2.0.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/68/bc/9104308fc285eb3e0b31b67688235db556cd5b0ef31d96f30e45f2e51cae/sphinxcontrib_qthelp-2.0.0.tar.gz", hash = "sha256:4fe7d0ac8fc171045be623aba3e2a8f613f8682731f9153bb2e40ece16b9bbab" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/27/83/859ecdd180cacc13b1f7e857abf8582a64552ea7a061057a6c716e790fce/sphinxcontrib_qthelp-2.0.0-py3-none-any.whl", hash = "sha256:b18a828cdba941ccd6ee8445dbe72ffa3ef8cbe7505d8cd1fa0d42d3f2d5f3eb" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "sphinxcontrib-serializinghtml"
version = "2.0.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/3b/44/6716b257b0aa6bfd51a1b31665d1c205fb12cb5ad56de752dfa15657de2f/sphinxcontrib_serializinghtml-2.0.0.tar.gz", hash = "sha256:e9d912827f872c029017a53f0ef2180b327c3f7fd23c87229f7a8e8b70031d4d" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/52/a7/d2782e4e3f77c8450f727ba74a8f12756d5ba823d81b941f1b04da9d033a/sphinxcontrib_serializinghtml-2.0.0-py3-none-any.whl", hash = "sha256:6e2cb0eef194e10c27ec0023bfeb25badbbb5868244cf5bc5bdc04e4464bf331" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
feat: add OceanBase doc engine (#11228) ### What problem does this PR solve? Add OceanBase doc engine. Close #5350 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-20 10:00:14 +08:00
[[package]]
name = "sqlalchemy"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "2.0.45"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
feat: add OceanBase doc engine (#11228) ### What problem does this PR solve? Add OceanBase doc engine. Close #5350 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-20 10:00:14 +08:00
dependencies = [
{ name = "greenlet", marker = "platform_machine == 'AMD64' or platform_machine == 'WIN32' or platform_machine == 'aarch64' or platform_machine == 'amd64' or platform_machine == 'ppc64le' or platform_machine == 'win32' or platform_machine == 'x86_64'" },
{ name = "typing-extensions" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/be/f9/5e4491e5ccf42f5d9cfc663741d261b3e6e1683ae7812114e7636409fcc6/sqlalchemy-2.0.45.tar.gz", hash = "sha256:1632a4bda8d2d25703fdad6363058d882541bdaaee0e5e3ddfa0cd3229efce88" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/6a/c8/7cc5221b47a54edc72a0140a1efa56e0a2730eefa4058d7ed0b4c4357ff8/sqlalchemy-2.0.45-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:fe187fc31a54d7fd90352f34e8c008cf3ad5d064d08fedd3de2e8df83eb4a1cf" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0e/50/80a8d080ac7d3d321e5e5d420c9a522b0aa770ec7013ea91f9a8b7d36e4a/sqlalchemy-2.0.45-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:672c45cae53ba88e0dad74b9027dddd09ef6f441e927786b05bec75d949fbb2e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/da/4c/13dab31266fc9904f7609a5dc308a2432a066141d65b857760c3bef97e69/sqlalchemy-2.0.45-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:470daea2c1ce73910f08caf10575676a37159a6d16c4da33d0033546bddebc9b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/74/04/891b5c2e9f83589de202e7abaf24cd4e4fa59e1837d64d528829ad6cc107/sqlalchemy-2.0.45-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:9c6378449e0940476577047150fd09e242529b761dc887c9808a9a937fe990c8" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f1/24/fc59e7f71b0948cdd4cff7a286210e86b0443ef1d18a23b0d83b87e4b1f7/sqlalchemy-2.0.45-cp313-cp313-win32.whl", hash = "sha256:4b6bec67ca45bc166c8729910bd2a87f1c0407ee955df110d78948f5b5827e8a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c0/c5/d17113020b2d43073412aeca09b60d2009442420372123b8d49cc253f8b8/sqlalchemy-2.0.45-cp313-cp313-win_amd64.whl", hash = "sha256:afbf47dc4de31fa38fd491f3705cac5307d21d4bb828a4f020ee59af412744ee" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3d/8d/bb40a5d10e7a5f2195f235c0b2f2c79b0bf6e8f00c0c223130a4fbd2db09/sqlalchemy-2.0.45-cp313-cp313t-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:83d7009f40ce619d483d26ac1b757dfe3167b39921379a8bd1b596cf02dab4a6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/75/a5/346128b0464886f036c039ea287b7332a410aa2d3fb0bb5d404cb8861635/sqlalchemy-2.0.45-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:d8a2ca754e5415cde2b656c27900b19d50ba076aa05ce66e2207623d3fe41f5a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/bf/e1/3ccb13c643399d22289c6a9786c1a91e3dcbb68bce4beb44926ac2c557bf/sqlalchemy-2.0.45-py3-none-any.whl", hash = "sha256:5225a288e4c8cc2308dbdd874edad6e7d0fd38eac1e9e5f23503425c8eee20d0" },
feat: add OceanBase doc engine (#11228) ### What problem does this PR solve? Add OceanBase doc engine. Close #5350 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-20 10:00:14 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "sqlglot"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "28.4.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/69/f1/a2b5174195448004f57092fb8d0e40466f9c650b9e660a7ee113d3de3e41/sqlglot-28.4.0.tar.gz", hash = "sha256:3ef93112e50a4427fbec2265a461595ee084a2fa80587d3b98be01d6a3699dfe" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/2d/a0/f2127b17b21ad9272d33152f57a8e1475a611599266b26f5149afea5c6c0/sqlglot-28.4.0-py3-none-any.whl", hash = "sha256:7861023184284d81bd3c502046ec6efacf31d17eb335ad10788e8aa1a06e19f0" },
bump infinity (#10422) ### What problem does this PR solve? bump infinity to v0.6.0-dev7 Needs https://github.com/infiniflow/infinity/pull/3016 ### Type of change - [x] Other (please describe): Infinity
2025-10-10 12:41:45 +08:00
]
[package.optional-dependencies]
rs = [
{ name = "sqlglotrs" },
]
[[package]]
name = "sqlglotrs"
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
version = "0.9.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/e0/c1/de7ee4729d49d15339717d6c4cc9aac06382c1161a8212dfdd266d51ffe5/sqlglotrs-0.9.0.tar.gz", hash = "sha256:72f61561d63607a8d88f5da608c11e21b2a57773ca631e6b89a4eed668da2db5" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/f3/27/6d42c98f2f33fc6dbbc7d669bf99ea6f7898d8bcd0aaf87aa1a4c96cc9c9/sqlglotrs-0.9.0-cp313-cp313-macosx_10_12_x86_64.whl", hash = "sha256:e2a5a697dbfc9cfee5434433a4d698a26df94277e0916bbfc25e1e72436cd0c0" },
{ url = "https://mirrors.aliyun.com/pypi/packages/50/53/d1f8f42ec14d69d8ba249036d83dcb4d6b51fe5b3ddb357499c737ae2a99/sqlglotrs-0.9.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:3a22d4064e923bbe07750f6e4b4b338e5b9fa0cbc2073bd503cc4b1c9280c2ac" },
{ url = "https://mirrors.aliyun.com/pypi/packages/52/e0/a2aa5e533427af4b64f9a630000cfee3cbbf877f58dcd79bb931963adf8a/sqlglotrs-0.9.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2fbf6f211d4b0d091855984279be7a9d57b89a43db07aeaf6cabee075c08ac80" },
{ url = "https://mirrors.aliyun.com/pypi/packages/65/fa/96bdaab19b7e8a09dec5a3bf3ec541569b23560a36df2d7d4b2bb910ac21/sqlglotrs-0.9.0-cp313-cp313-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:a34227d5f012d379ff7e3d87f2b59c065456470c9d9a31971074942a8bd02ac8" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f8/48/0813a8bca74477115aee180a6570b4d67d74b2d08997f3a1beb4f704dd89/sqlglotrs-0.9.0-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:a681b3d5ddee941d862fcd6c14fc1ce54d30fcfe06c5910689dc3d609b26cdf4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3d/da/dc7cb16c7a958d275695c3761cd8203f50149902085f71b7d86dad981241/sqlglotrs-0.9.0-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:44f4d6e6b153a7397aaa19171549ad206dccb8fa43f145daea708ced9c30f39d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8c/e4/9f6c340a5ff7c9f45bc2c972b142e51d29a80a7e02619891a08faac2ecfa/sqlglotrs-0.9.0-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3f3f6d7b64d443cf7a030a441a0a4de5988878e64784e72bf7fabfb8d0d9cc0b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/25/9f/a888afe6da474ef8fb76cbddd0123bbea50e0c67c4bfcb4863baf0e4e6c7/sqlglotrs-0.9.0-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:6889fc464da28f0db96ff1daf44ac02a5b97e1fd3c3ca78419248e18b2b07485" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ba/86/bc4cdabc597f099e4505803633f0e7a73e9a907a8afc8a005d9afa218c1f/sqlglotrs-0.9.0-cp313-cp313-win32.whl", hash = "sha256:dd3ca532c088b747208dd3fa67aa2d0cbf9df7a7258718085bbe8c21ecbf3482" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0b/d5/6596837933e702e96677df014891efb5eb26436c900e04f7712e7048c75d/sqlglotrs-0.9.0-cp313-cp313-win_amd64.whl", hash = "sha256:bb9d191f98eb7b7240cc45a730cd1d172fa79e2d195e3c864832476d594c51ae" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
[[package]]
name = "srsly"
version = "2.5.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
dependencies = [
{ name = "catalogue" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/2b/db/f794f219a6c788b881252d2536a8c4a97d2bdaadc690391e1cb53d123d71/srsly-2.5.3.tar.gz", hash = "sha256:08f98dbecbff3a31466c4ae7c833131f59d3655a0ad8ac749e6e2c149e2b0680" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/9d/5c/12901e3794f4158abc6da750725aad6c2afddb1e4227b300fe7c71f66957/srsly-2.5.3-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:e67b6bbacbfadea5e100266d2797f2d4cec9883ea4dc84a5537673850036a8d8" },
{ url = "https://mirrors.aliyun.com/pypi/packages/04/61/181c26370995f96f56f1b64b801e3ca1e0d703fc36506ae28606d62369fb/srsly-2.5.3-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:348c231b4477d8fe86603131d0f166d2feac9c372704dfc4398be71cc5b6fb07" },
{ url = "https://mirrors.aliyun.com/pypi/packages/77/c6/35876c78889f8ffe11ed3521644e666c3aef20ea31527b70f47456cf35c2/srsly-2.5.3-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:b0938c2978c91ae1ef9c1f2ba35abb86330e198fb23469e356eba311e02233ee" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3e/da/40b71ca9906c8eb8f8feb6ac11d33dad458c85a56e1de764b96d402168a0/srsly-2.5.3-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:5f6a837954429ecbe6dcdd27390d2fb4c7d01a3f99c9ffcf9ce66b2a6dd1b738" },
{ url = "https://mirrors.aliyun.com/pypi/packages/dc/14/c0dd30cc8b93ce8137ff4766f743c882440ce49195fffc5d50eaeef311a6/srsly-2.5.3-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:3576c125c486ce2958c2047e8858fe3cfc9ea877adfa05203b0986f9badee355" },
{ url = "https://mirrors.aliyun.com/pypi/packages/08/f3/34354f183d8faafc631585571224b54d1b4b67e796972c36519c074ca355/srsly-2.5.3-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:5fb59c42922e095d1ea36085c55bc16e2adb06a7bfe57b24d381e0194ae699f2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a4/d9/5531f8a19492060b4e76e4ab06aca6f096fb5128fe18cc813d1772daf653/srsly-2.5.3-cp313-cp313-win_amd64.whl", hash = "sha256:111805927f05f5db440aeeacb85ce43da0b19ce7b2a09567a9ef8d30f3cc4d83" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8e/8a/62fb7a971eca29e12f03fb9ddacb058548c14d33e5b5675ff0f85839cc7b/srsly-2.5.3-cp313-cp313-win_arm64.whl", hash = "sha256:0f106b0a700ab56e4a7c431b0f1444009ab6cb332edc7bbf6811c2a43f4722cb" },
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
]
Feat: mcp server (#7084) ### What problem does this PR solve? Add MCP support with a client example. Issue link: #4344 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-04-21 09:43:20 +08:00
[[package]]
name = "sse-starlette"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "3.1.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: mcp server (#7084) ### What problem does this PR solve? Add MCP support with a client example. Issue link: #4344 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-04-21 09:43:20 +08:00
dependencies = [
{ name = "anyio" },
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
{ name = "starlette" },
Feat: mcp server (#7084) ### What problem does this PR solve? Add MCP support with a client example. Issue link: #4344 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-04-21 09:43:20 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/da/34/f5df66cb383efdbf4f2db23cabb27f51b1dcb737efaf8a558f6f1d195134/sse_starlette-3.1.2.tar.gz", hash = "sha256:55eff034207a83a0eb86de9a68099bd0157838f0b8b999a1b742005c71e33618" }
Feat: mcp server (#7084) ### What problem does this PR solve? Add MCP support with a client example. Issue link: #4344 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-04-21 09:43:20 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/b7/95/8c4b76eec9ae574474e5d2997557cebf764bcd3586458956c30631ae08f4/sse_starlette-3.1.2-py3-none-any.whl", hash = "sha256:cd800dd349f4521b317b9391d3796fa97b71748a4da9b9e00aafab32dda375c8" },
Feat: mcp server (#7084) ### What problem does this PR solve? Add MCP support with a client example. Issue link: #4344 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-04-21 09:43:20 +08:00
]
[[package]]
name = "starlette"
build(deps): bump starlette from 0.51.0 to 1.3.1 (#16089) Bumps [starlette](https://github.com/Kludex/starlette) from 0.51.0 to 1.3.1. <details> <summary>Release notes</summary> <p><em>Sourced from <a href="https://github.com/Kludex/starlette/releases">starlette's releases</a>.</em></p> <blockquote> <h2>Version 1.3.1</h2> <h2>What's Changed</h2> <ul> <li>Use <code>StarletteDeprecationWarning</code> instead of <code>DeprecationWarning</code> by <a href="https://github.com/Kludex"><code>@​Kludex</code></a> in <a href="https://redirect.github.com/Kludex/starlette/pull/3119">Kludex/starlette#3119</a></li> <li>Enforce <code>max_fields</code> and <code>max_part_size</code> in <code>FormParser</code> by <a href="https://github.com/Kludex"><code>@​Kludex</code></a> in <a href="https://redirect.github.com/Kludex/starlette/pull/3329">Kludex/starlette#3329</a></li> <li>Enforce <code>FormParser</code> limits in parser callbacks by <a href="https://github.com/Kludex"><code>@​Kludex</code></a> in <a href="https://redirect.github.com/Kludex/starlette/pull/3331">Kludex/starlette#3331</a></li> </ul> <p><strong>Full Changelog</strong>: <a href="https://github.com/Kludex/starlette/compare/1.3.0...1.3.1">https://github.com/Kludex/starlette/compare/1.3.0...1.3.1</a></p> <h2>Version 1.3.0</h2> <h2>What's Changed</h2> <ul> <li>Clamp oversized suffix ranges in <code>FileResponse</code> by <a href="https://github.com/jiyujie2006"><code>@​jiyujie2006</code></a> in <a href="https://redirect.github.com/Kludex/starlette/pull/3307">Kludex/starlette#3307</a></li> <li>Catch <code>OSError</code> alongside <code>MultiPartException</code> when closing temp files by <a href="https://github.com/N3XT3R1337"><code>@​N3XT3R1337</code></a> in <a href="https://redirect.github.com/Kludex/starlette/pull/3191">Kludex/starlette#3191</a></li> <li>Add <code>httpx2</code> to the <code>full</code> extra by <a href="https://github.com/Kludex"><code>@​Kludex</code></a> in <a href="https://redirect.github.com/Kludex/starlette/pull/3323">Kludex/starlette#3323</a></li> <li>Adjust testclient typing and warnings by <a href="https://github.com/waketzheng"><code>@​waketzheng</code></a> in <a href="https://redirect.github.com/Kludex/starlette/pull/3322">Kludex/starlette#3322</a></li> <li>Fix IndexError in URL.replace() on a URL with no authority by <a href="https://github.com/LeSingh1"><code>@​LeSingh1</code></a> in <a href="https://redirect.github.com/Kludex/starlette/pull/3317">Kludex/starlette#3317</a></li> <li>Annotate URLPath protocol parameter with Literal by <a href="https://github.com/Chang-LeHung"><code>@​Chang-LeHung</code></a> in <a href="https://redirect.github.com/Kludex/starlette/pull/3285">Kludex/starlette#3285</a></li> <li>avoid collapsing exception groups from user code by <a href="https://github.com/graingert"><code>@​graingert</code></a> in <a href="https://redirect.github.com/Kludex/starlette/pull/2830">Kludex/starlette#2830</a></li> <li>Use <code>removeprefix</code> to strip weak ETag indicator in <code>is_not_modified</code> by <a href="https://github.com/gnosyslambda"><code>@​gnosyslambda</code></a> in <a href="https://redirect.github.com/Kludex/starlette/pull/3193">Kludex/starlette#3193</a></li> <li>Build <code>request.url</code> from structured components by <a href="https://github.com/Kludex"><code>@​Kludex</code></a> in <a href="https://redirect.github.com/Kludex/starlette/pull/3326">Kludex/starlette#3326</a></li> </ul> <h2>New Contributors</h2> <ul> <li><a href="https://github.com/jiyujie2006"><code>@​jiyujie2006</code></a> made their first contribution in <a href="https://redirect.github.com/Kludex/starlette/pull/3307">Kludex/starlette#3307</a></li> <li><a href="https://github.com/N3XT3R1337"><code>@​N3XT3R1337</code></a> made their first contribution in <a href="https://redirect.github.com/Kludex/starlette/pull/3191">Kludex/starlette#3191</a></li> <li><a href="https://github.com/leestana01"><code>@​leestana01</code></a> made their first contribution in <a href="https://redirect.github.com/Kludex/starlette/pull/3319">Kludex/starlette#3319</a></li> <li><a href="https://github.com/LeSingh1"><code>@​LeSingh1</code></a> made their first contribution in <a href="https://redirect.github.com/Kludex/starlette/pull/3317">Kludex/starlette#3317</a></li> <li><a href="https://github.com/EmmanuelNiyonshuti"><code>@​EmmanuelNiyonshuti</code></a> made their first contribution in <a href="https://redirect.github.com/Kludex/starlette/pull/3204">Kludex/starlette#3204</a></li> <li><a href="https://github.com/Chang-LeHung"><code>@​Chang-LeHung</code></a> made their first contribution in <a href="https://redirect.github.com/Kludex/starlette/pull/3285">Kludex/starlette#3285</a></li> <li><a href="https://github.com/gnosyslambda"><code>@​gnosyslambda</code></a> made their first contribution in <a href="https://redirect.github.com/Kludex/starlette/pull/3193">Kludex/starlette#3193</a></li> </ul> <p><strong>Full Changelog</strong>: <a href="https://github.com/Kludex/starlette/compare/1.2.1...1.3.0">https://github.com/Kludex/starlette/compare/1.2.1...1.3.0</a></p> <h2>Version 1.2.1</h2> <h2>What's Changed</h2> <ul> <li>Use <code>httpx2</code> for type checking in the <code>testclient</code> module by <a href="https://github.com/leifwar"><code>@​leifwar</code></a> in <a href="https://redirect.github.com/Kludex/starlette/pull/3304">Kludex/starlette#3304</a></li> <li>Add assert error for requires() when request param is not Request type by <a href="https://github.com/KeeganOP"><code>@​KeeganOP</code></a> in <a href="https://redirect.github.com/Kludex/starlette/pull/3298">Kludex/starlette#3298</a></li> </ul> <h2>New Contributors</h2> <ul> <li><a href="https://github.com/leifwar"><code>@​leifwar</code></a> made their first contribution in <a href="https://redirect.github.com/Kludex/starlette/pull/3304">Kludex/starlette#3304</a></li> <li><a href="https://github.com/diskeu"><code>@​diskeu</code></a> made their first contribution in <a href="https://redirect.github.com/Kludex/starlette/pull/3243">Kludex/starlette#3243</a></li> <li><a href="https://github.com/KeeganOP"><code>@​KeeganOP</code></a> made their first contribution in <a href="https://redirect.github.com/Kludex/starlette/pull/3298">Kludex/starlette#3298</a></li> </ul> <p><strong>Full Changelog</strong>: <a href="https://github.com/Kludex/starlette/compare/1.2.0...1.2.1">https://github.com/Kludex/starlette/compare/1.2.0...1.2.1</a></p> <h2>Version 1.2.0</h2> <h2>What's Changed</h2> <ul> <li>Support httpx2 in the test client by <a href="https://github.com/Kludex"><code>@​Kludex</code></a> in <a href="https://redirect.github.com/Kludex/starlette/pull/3291">Kludex/starlette#3291</a></li> </ul> <p><strong>Full Changelog</strong>: <a href="https://github.com/Kludex/starlette/compare/1.1.0...1.2.0">https://github.com/Kludex/starlette/compare/1.1.0...1.2.0</a></p> <h2>Version 1.1.0</h2> <!-- raw HTML omitted --> </blockquote> <p>... (truncated)</p> </details> <details> <summary>Changelog</summary> <p><em>Sourced from <a href="https://github.com/Kludex/starlette/blob/main/docs/release-notes.md">starlette's changelog</a>.</em></p> <blockquote> <h2>1.3.1 (June 12, 2026)</h2> <h4>Fixed</h4> <ul> <li>Enforce <code>max_fields</code> and <code>max_part_size</code> in <code>FormParser</code> <a href="https://redirect.github.com/encode/starlette/pull/3329">#3329</a>.</li> <li>Enforce <code>FormParser</code> limits in parser callbacks <a href="https://redirect.github.com/encode/starlette/pull/3331">#3331</a>.</li> </ul> <h2>1.3.0 (June 11, 2026)</h2> <h4>Added</h4> <ul> <li>Add <code>httpx2</code> to the <code>full</code> extra <a href="https://redirect.github.com/encode/starlette/pull/3323">#3323</a>.</li> <li>Annotate the <code>URLPath</code> <code>protocol</code> parameter with <code>Literal</code> <a href="https://redirect.github.com/encode/starlette/pull/3285">#3285</a>.</li> </ul> <h4>Fixed</h4> <ul> <li>Build <code>request.url</code> from structured components <a href="https://redirect.github.com/encode/starlette/pull/3326">#3326</a>.</li> <li>Clamp oversized suffix ranges in <code>FileResponse</code> <a href="https://redirect.github.com/encode/starlette/pull/3307">#3307</a>.</li> <li>Catch <code>OSError</code> alongside <code>MultiPartException</code> when closing temp files <a href="https://redirect.github.com/encode/starlette/pull/3191">#3191</a>.</li> <li>Avoid collapsing exception groups raised from user code <a href="https://redirect.github.com/encode/starlette/pull/2830">#2830</a>.</li> <li>Use <code>removeprefix</code> to strip the weak <code>ETag</code> indicator in <code>is_not_modified</code> <a href="https://redirect.github.com/encode/starlette/pull/3193">#3193</a>.</li> <li>Fix <code>IndexError</code> in <code>URL.replace()</code> on a URL with no authority <a href="https://redirect.github.com/encode/starlette/pull/3317">#3317</a>.</li> <li>Adjust <code>testclient</code> typing and warnings <a href="https://redirect.github.com/encode/starlette/pull/3322">#3322</a>.</li> </ul> <h2>1.2.1 (May 31, 2026)</h2> <h4>Fixed</h4> <ul> <li>Use <code>httpx2</code> for type checking in the <code>testclient</code> module <a href="https://redirect.github.com/encode/starlette/pull/3304">#3304</a>.</li> <li>Add assert error for <code>requires()</code> when the request parameter is not a <code>Request</code> type <a href="https://redirect.github.com/encode/starlette/pull/3298">#3298</a>.</li> </ul> <h2>1.2.0 (May 28, 2026)</h2> <h4>Added</h4> <ul> <li>Support httpx2 in the test client <a href="https://redirect.github.com/encode/starlette/pull/3291">#3291</a>.</li> </ul> <h2>1.1.0 (May 23, 2026)</h2> <h4>Added</h4> <ul> <li>Use <code>&quot;application/octet-stream&quot;</code> as the <code>FileResponse</code> media type fallback <a href="https://redirect.github.com/encode/starlette/pull/3283">#3283</a>.</li> </ul> <h4>Fixed</h4> <ul> <li>Only dispatch standard HTTP verbs in <code>HTTPEndpoint</code> <a href="https://redirect.github.com/encode/starlette/pull/3286">#3286</a>.</li> <li>Reject absolute paths in <code>StaticFiles.lookup_path</code> <a href="https://redirect.github.com/encode/starlette/pull/3287">#3287</a>.</li> </ul> <h2>1.0.1 (May 21, 2026)</h2> <!-- raw HTML omitted --> </blockquote> <p>... (truncated)</p> </details> <details> <summary>Commits</summary> <ul> <li><a href="https://github.com/Kludex/starlette/commit/8ebffd0678570ddd5d5bb11c6f3c3c7fd4682ab9"><code>8ebffd0</code></a> Version 1.3.1 (<a href="https://redirect.github.com/Kludex/starlette/issues/3330">#3330</a>)</li> <li><a href="https://github.com/Kludex/starlette/commit/25b8e179d8d7ed86769c02f648772dd5fb43dc3c"><code>25b8e17</code></a> Enforce <code>FormParser</code> limits in parser callbacks (<a href="https://redirect.github.com/Kludex/starlette/issues/3331">#3331</a>)</li> <li><a href="https://github.com/Kludex/starlette/commit/dba1c4babc4f99ad2622bb913d87045775dda735"><code>dba1c4b</code></a> Enforce <code>max_fields</code> and <code>max_part_size</code> in <code>FormParser</code> (<a href="https://redirect.github.com/Kludex/starlette/issues/3329">#3329</a>)</li> <li><a href="https://github.com/Kludex/starlette/commit/45e51dcf99f3a270b0bcec1aec5410b4345863a9"><code>45e51dc</code></a> Use <code>StarletteDeprecationWarning</code> instead of <code>DeprecationWarning</code> (<a href="https://redirect.github.com/Kludex/starlette/issues/3119">#3119</a>)</li> <li><a href="https://github.com/Kludex/starlette/commit/5f8610c386e13de1d80d36efa961e1486a1d2d01"><code>5f8610c</code></a> Version 1.3.0 (<a href="https://redirect.github.com/Kludex/starlette/issues/3327">#3327</a>)</li> <li><a href="https://github.com/Kludex/starlette/commit/167b5850e809f38b27fbfed62d58bf6442855975"><code>167b585</code></a> Build <code>request.url</code> from structured components (<a href="https://redirect.github.com/Kludex/starlette/issues/3326">#3326</a>)</li> <li><a href="https://github.com/Kludex/starlette/commit/37309255b4c1b9c381a2d24a1eaf83100984a16a"><code>3730925</code></a> Use <code>removeprefix</code> to strip weak ETag indicator in <code>is_not_modified</code> (<a href="https://redirect.github.com/Kludex/starlette/issues/3193">#3193</a>)</li> <li><a href="https://github.com/Kludex/starlette/commit/e6f7ad1ab85efb27ab7910d8007b3f4531f7b083"><code>e6f7ad1</code></a> avoid collapsing exception groups from user code (<a href="https://redirect.github.com/Kludex/starlette/issues/2830">#2830</a>)</li> <li><a href="https://github.com/Kludex/starlette/commit/115228fcdca0e0ef5bf4a95a40ddce5a9fced428"><code>115228f</code></a> Annotate URLPath protocol parameter with Literal (<a href="https://redirect.github.com/Kludex/starlette/issues/3285">#3285</a>)</li> <li><a href="https://github.com/Kludex/starlette/commit/113f193a34353c9153857028c1074351d22fad07"><code>113f193</code></a> docs: replace inline ASGI server list with link to canonical implemen… (<a href="https://redirect.github.com/Kludex/starlette/issues/3204">#3204</a>)</li> <li>Additional commits viewable in <a href="https://github.com/Kludex/starlette/compare/0.51.0...1.3.1">compare view</a></li> </ul> </details> <br /> [![Dependabot compatibility score](https://dependabot-badges.githubapp.com/badges/compatibility_score?dependency-name=starlette&package-manager=uv&previous-version=0.51.0&new-version=1.3.1)](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores) Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting `@dependabot rebase`. [//]: # (dependabot-automerge-start) [//]: # (dependabot-automerge-end) --- <details> <summary>Dependabot commands and options</summary> <br /> You can trigger Dependabot actions by commenting on this PR: - `@dependabot rebase` will rebase this PR - `@dependabot recreate` will recreate this PR, overwriting any edits that have been made to it - `@dependabot show <dependency name> ignore conditions` will show all of the ignore conditions of the specified dependency - `@dependabot ignore this major version` will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself) - `@dependabot ignore this minor version` will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself) - `@dependabot ignore this dependency` will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself) You can disable automated security fix PRs for this repo from the [Security Alerts page](https://github.com/infiniflow/ragflow/network/alerts). </details> Co-authored-by: Yingfeng Zhang <yingfeng.zhang@gmail.com>
2026-06-16 19:24:45 +08:00
version = "1.3.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: mcp server (#7084) ### What problem does this PR solve? Add MCP support with a client example. Issue link: #4344 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-04-21 09:43:20 +08:00
dependencies = [
{ name = "anyio" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/eb/e3/7c1dc7381d9f8ab7d854328ebfa884e62cb3f3d8549ddfd37c7814f42afa/starlette-1.3.1.tar.gz", hash = "sha256:05d0213193f2fbaae60e2ecb593b4add4262ad4e46536b54abe36f11a71724e0" }
Feat: mcp server (#7084) ### What problem does this PR solve? Add MCP support with a client example. Issue link: #4344 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-04-21 09:43:20 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ec/bb/2799cc2ede3ed41131f8975621e7213dfc7ef4acbbaadfa440f32500c370/starlette-1.3.1-py3-none-any.whl", hash = "sha256:c7372aae11c3c3f26a42df7bd626cec2f47d03483d261d369516a615a53714c6" },
Feat: mcp server (#7084) ### What problem does this PR solve? Add MCP support with a client example. Issue link: #4344 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-04-21 09:43:20 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "statsmodels"
Fix: aspose-slide issue. (#11935) ### Type of change - [x] Bug Fix (non-breaking change which fixes an issue)
2025-12-12 20:16:18 +08:00
version = "0.14.6"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "numpy" },
{ name = "packaging" },
{ name = "pandas" },
{ name = "patsy" },
Bump infinity to v0.6.11. Requires python>=3.11 (#11814) ### What problem does this PR solve? Bump infinity to v0.6.11. Requires python>=3.11 ### Type of change - [x] Refactoring
2025-12-09 16:23:37 +08:00
{ name = "scipy" },
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/0d/81/e8d74b34f85285f7335d30c5e3c2d7c0346997af9f3debf9a0a9a63de184/statsmodels-0.14.6.tar.gz", hash = "sha256:4d17873d3e607d398b85126cd4ed7aad89e4e9d89fc744cdab1af3189a996c2a" }
Security: do not use litellm 1.82.7 and 1.82.8 (#13768) ### What problem does this PR solve? See [issue](https://github.com/BerriAI/litellm/issues/24518) from Litellm. Upgraded from `1.81.15` to `1.82.6`, so RAGFlow is safe as always. ### Type of change - [x] Security Co-authored-by: Jin Hai <haijin.chn@gmail.com> Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
2026-03-25 22:39:33 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/81/59/a5aad5b0cc266f5be013db8cde563ac5d2a025e7efc0c328d83b50c72992/statsmodels-0.14.6-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:47ee7af083623d2091954fa71c7549b8443168f41b7c5dce66510274c50fd73e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/53/dd/d8cfa7922fc6dc3c56fa6c59b348ea7de829a94cd73208c6f8202dd33f17/statsmodels-0.14.6-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:aa60d82e29fcd0a736e86feb63a11d2380322d77a9369a54be8b0965a3985f71" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ee/77/0ec96803eba444efd75dba32f2ef88765ae3e8f567d276805391ec2c98c6/statsmodels-0.14.6-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:89ee7d595f5939cc20bf946faedcb5137d975f03ae080f300ebb4398f16a5bd4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/10/b9/fd41f1f6af13a1a1212a06bb377b17762feaa6d656947bf666f76300fc05/statsmodels-0.14.6-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:730f3297b26749b216a06e4327fe0be59b8d05f7d594fb6caff4287b69654589" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ee/0f/a6900e220abd2c69cd0a07e3ad26c71984be6061415a60e0f17b152ecf08/statsmodels-0.14.6-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:f1c08befa85e93acc992b72a390ddb7bd876190f1360e61d10cf43833463bc9c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/98/08/b79f0c614f38e566eebbdcff90c0bcacf3c6ba7a5bbb12183c09c29ca400/statsmodels-0.14.6-cp313-cp313-win_amd64.whl", hash = "sha256:8021271a79f35b842c02a1794465a651a9d06ec2080f76ebc3b7adce77d08233" },
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
]
[[package]]
name = "stone"
version = "3.3.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
dependencies = [
{ name = "ply" },
{ name = "six" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/99/6f/ef25bbc1aefeb9c905d527f1d3cd3f41f22f40566d33001b8bb14ae0cdaf/stone-3.3.1.tar.gz", hash = "sha256:4ef0397512f609757975f7ec09b35639d72ba7e3e17ce4ddf399578346b4cb50" }
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/5c/92/d0c83f63d3518e5f0b8a311937c31347349ec9a47b209ddc17f7566f58fc/stone-3.3.1-py3-none-any.whl", hash = "sha256:e15866fad249c11a963cce3bdbed37758f2e88c8ff4898616bc0caeb1e216047" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "sympy"
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
version = "1.14.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "mpmath" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/83/d3/803453b36afefb7c2bb238361cd4ae6125a569b4db67cd9e79846ba2d68c/sympy-1.14.0.tar.gz", hash = "sha256:d3d3fe8df1e5a0b42f0e7bdf50541697dbe7d23746e894990c030e2b05e72517" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/a2/09/77d55d46fd61b4a135c444fc97158ef34a095e5681d0a6c10b75bf356191/sympy-1.14.0-py3-none-any.whl", hash = "sha256:e091cc3e99d2141a0ba2847328f5479b05d94a6635cb96148ccb3f34671bd8f5" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
[[package]]
name = "tablestore"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
version = "6.4.6"
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
dependencies = [
{ name = "aiohttp" },
{ name = "certifi" },
{ name = "crc32c" },
{ name = "flatbuffers" },
{ name = "future" },
{ name = "numpy" },
{ name = "protobuf" },
{ name = "six" },
{ name = "urllib3" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/90/1b/03bee654a922a818a12df211bbf31b724773082db356d964579e4b004f5b/tablestore-6.4.6.tar.gz", hash = "sha256:b0f3fd864380b0448c0dfc648dc7bf28e41c35afab8ddf6ee75cf43e480a1abc" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/ab/24/6cd95c15123a6b2a17fbc013e78a4f6119df987a2004f8996768f060a353/tablestore-6.4.6-py3-none-any.whl", hash = "sha256:060f24cff09c4e2912e01adaf1be3b9a69d609f3f82e83f7c979136357e7e8fd" },
]
[[package]]
name = "tablestore-agent-storage"
version = "1.0.6"
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
dependencies = [
{ name = "oss2" },
{ name = "tablestore" },
]
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/c5/42/e94edd2ab55adbe8f986906a7005e1944ce7fd891a66e5b23e0a9f67d6d4/tablestore_agent_storage-1.0.6.tar.gz", hash = "sha256:bbf5458ff8d249aa20845279846958605273a452c6eae88e15bcc86dda429a83" }
Security: do not use litellm 1.82.7 and 1.82.8 (#13768) ### What problem does this PR solve? See [issue](https://github.com/BerriAI/litellm/issues/24518) from Litellm. Upgraded from `1.81.15` to `1.82.6`, so RAGFlow is safe as always. ### Type of change - [x] Security Co-authored-by: Jin Hai <haijin.chn@gmail.com> Co-authored-by: Yingfeng <yingfeng.zhang@gmail.com>
2026-03-25 22:39:33 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/b8/51/8cc6b37837ee606bc71615bb95795815fbf5a6fed24acfabc65cb6d2dbda/tablestore_agent_storage-1.0.6-py3-none-any.whl", hash = "sha256:e8abf19679a6bde110fa77e01d37e25ee4aada3f64f7a50fcb3429cd0a79718b" },
feat: Implement pluggable multi-provider sandbox architecture (#12820) ## Summary Implement a flexible sandbox provider system supporting both self-managed (Docker) and SaaS (Aliyun Code Interpreter) backends for secure code execution in agent workflows. **Key Changes:** - ✅ Aliyun Code Interpreter provider using official `agentrun-sdk>=0.0.16` - ✅ Self-managed provider with gVisor (runsc) security - ✅ Arguments parameter support for dynamic code execution - ✅ Database-only configuration (removed fallback logic) - ✅ Configuration scripts for quick setup Issue #12479 ## Features ### 🔌 Provider Abstraction Layer **1. Self-Managed Provider** (`agent/sandbox/providers/self_managed.py`) - Wraps existing executor_manager HTTP API - gVisor (runsc) for secure container isolation - Configurable pool size, timeout, retry logic - Languages: Python, Node.js, JavaScript - ⚠️ **Requires**: gVisor installation, Docker, base images **2. Aliyun Code Interpreter** (`agent/sandbox/providers/aliyun_codeinterpreter.py`) - SaaS integration using official agentrun-sdk - Serverless microVM execution with auto-authentication - Hard timeout: 30 seconds max - Credentials: `AGENTRUN_ACCESS_KEY_ID`, `AGENTRUN_ACCESS_KEY_SECRET`, `AGENTRUN_ACCOUNT_ID`, `AGENTRUN_REGION` - Automatically wraps code to call `main()` function **3. E2B Provider** (`agent/sandbox/providers/e2b.py`) - Placeholder for future integration ### ⚙️ Configuration System - `conf/system_settings.json`: Default provider = `aliyun_codeinterpreter` - `agent/sandbox/client.py`: Enforces database-only configuration - Admin UI: `/admin/sandbox-settings` - Configuration validation via `validate_config()` method - Health checks for all providers ### 🎯 Key Capabilities **Arguments Parameter Support:** All providers support passing arguments to `main()` function: ```python # User code def main(name: str, count: int) -> dict: return {"message": f"Hello {name}!" * count} # Executed with: arguments={"name": "World", "count": 3} # Result: {"message": "Hello World!Hello World!Hello World!"} ``` **Self-Describing Providers:** Each provider implements `get_config_schema()` returning form configuration for Admin UI **Error Handling:** Structured `ExecutionResult` with stdout, stderr, exit_code, execution_time ## Configuration Scripts Two scripts for quick Aliyun sandbox setup: **Shell Script (requires jq):** ```bash source scripts/configure_aliyun_sandbox.sh ``` **Python Script (interactive):** ```bash python3 scripts/configure_aliyun_sandbox.py ``` ## Testing ```bash # Unit tests uv run pytest agent/sandbox/tests/test_providers.py -v # Aliyun provider tests uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter.py -v # Integration tests (requires credentials) uv run pytest agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py -v # Quick SDK validation python3 agent/sandbox/tests/verify_sdk.py ``` **Test Coverage:** - 30 unit tests for provider abstraction - Provider-specific tests for Aliyun - Integration tests with real API - Security tests for executor_manager ## Documentation - `docs/develop/sandbox_spec.md` - Complete architecture specification - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration from legacy sandbox - `agent/sandbox/tests/QUICKSTART.md` - Quick start guide - `agent/sandbox/tests/README.md` - Testing documentation ## Breaking Changes ⚠️ **Migration Required:** 1. **Directory Move**: `sandbox/` → `agent/sandbox/` - Update imports: `from sandbox.` → `from agent.sandbox.` 2. **Mandatory Configuration**: - SystemSettings must have `sandbox.provider_type` configured - Removed fallback default values - Configuration must exist in database (from `conf/system_settings.json`) 3. **Aliyun Credentials**: - Requires `AGENTRUN_*` environment variables (not `ALIYUN_*`) - `AGENTRUN_ACCOUNT_ID` is now required (Aliyun primary account ID) 4. **Self-Managed Provider**: - gVisor (runsc) must be installed for security - Install: `go install gvisor.dev/gvisor/runsc@latest` ## Database Schema Changes ```python # SystemSettings.value: CharField → TextField api/db/db_models.py: Changed for unlimited config length # SystemSettingsService.get_by_name(): Fixed query precision api/db/services/system_settings_service.py: startswith → exact match ``` ## Files Changed ### Backend (Python) - `agent/sandbox/providers/base.py` - SandboxProvider ABC interface - `agent/sandbox/providers/manager.py` - ProviderManager - `agent/sandbox/providers/self_managed.py` - Self-managed provider - `agent/sandbox/providers/aliyun_codeinterpreter.py` - Aliyun provider - `agent/sandbox/providers/e2b.py` - E2B provider (placeholder) - `agent/sandbox/client.py` - Unified client (enforces DB-only config) - `agent/tools/code_exec.py` - Updated to use provider system - `admin/server/services.py` - SandboxMgr with registry & validation - `admin/server/routes.py` - 5 sandbox API endpoints - `conf/system_settings.json` - Default: aliyun_codeinterpreter - `api/db/db_models.py` - TextField for SystemSettings.value - `api/db/services/system_settings_service.py` - Exact match query ### Frontend (TypeScript/React) - `web/src/pages/admin/sandbox-settings.tsx` - Settings UI - `web/src/services/admin-service.ts` - Sandbox service functions - `web/src/services/admin.service.d.ts` - Type definitions - `web/src/utils/api.ts` - Sandbox API endpoints ### Documentation - `docs/develop/sandbox_spec.md` - Architecture spec - `agent/sandbox/tests/MIGRATION_GUIDE.md` - Migration guide - `agent/sandbox/tests/QUICKSTART.md` - Quick start - `agent/sandbox/tests/README.md` - Testing guide ### Configuration Scripts - `scripts/configure_aliyun_sandbox.sh` - Shell script (jq) - `scripts/configure_aliyun_sandbox.py` - Python script ### Tests - `agent/sandbox/tests/test_providers.py` - 30 unit tests - `agent/sandbox/tests/test_aliyun_codeinterpreter.py` - Provider tests - `agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py` - Integration tests - `agent/sandbox/tests/verify_sdk.py` - SDK validation ## Architecture ``` Admin UI → Admin API → SandboxMgr → ProviderManager → [SelfManaged|Aliyun|E2B] ↓ SystemSettings ``` ## Usage ### 1. Configure Provider **Via Admin UI:** 1. Navigate to `/admin/sandbox-settings` 2. Select provider (Aliyun Code Interpreter / Self-Managed) 3. Fill in configuration 4. Click "Test Connection" to verify 5. Click "Save" to apply **Via Configuration Scripts:** ```bash # Aliyun provider export AGENTRUN_ACCESS_KEY_ID="xxx" export AGENTRUN_ACCESS_KEY_SECRET="yyy" export AGENTRUN_ACCOUNT_ID="zzz" export AGENTRUN_REGION="cn-shanghai" source scripts/configure_aliyun_sandbox.sh ``` ### 2. Restart Service ```bash cd docker docker compose restart ragflow-server ``` ### 3. Execute Code in Agent ```python from agent.sandbox.client import execute_code result = execute_code( code='def main(name: str) -> dict: return {"message": f"Hello {name}!"}', language="python", timeout=30, arguments={"name": "World"} ) print(result.stdout) # {"message": "Hello World!"} ``` ## Troubleshooting ### "Container pool is busy" (Self-Managed) - **Cause**: Pool exhausted (default: 1 container in `.env`) - **Fix**: Increase `SANDBOX_EXECUTOR_MANAGER_POOL_SIZE` to 5+ ### "Sandbox provider type not configured" - **Cause**: Database missing configuration - **Fix**: Run config script or set via Admin UI ### "gVisor not found" - **Cause**: runsc not installed - **Fix**: `go install gvisor.dev/gvisor/runsc@latest && sudo cp ~/go/bin/runsc /usr/local/bin/` ### Aliyun authentication errors - **Cause**: Wrong environment variable names - **Fix**: Use `AGENTRUN_*` prefix (not `ALIYUN_*`) ## Checklist - [x] All tests passing (30 unit tests + integration tests) - [x] Documentation updated (spec, migration guide, quickstart) - [x] Type definitions added (TypeScript) - [x] Admin UI implemented - [x] Configuration validation - [x] Health checks implemented - [x] Error handling with structured results - [x] Breaking changes documented - [x] Configuration scripts created - [x] gVisor requirements documented Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
2026-01-28 13:28:21 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "tabulate"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "0.9.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ec/fe/802052aecb21e3797b8f7902564ab6ea0d60ff8ca23952079064155d1ae1/tabulate-0.9.0.tar.gz", hash = "sha256:0095b12bf5966de529c0feb1fa08671671b3368eec77d7ef7ab114be2c068b3c" }
Feat: Alter flask to Quart for async API serving. (#11275) ### What problem does this PR solve? #11277 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-18 17:05:16 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/40/44/4a5f08c96eb108af5cb50b41f76142f0afa346dfa99d5296fe7202a11854/tabulate-0.9.0-py3-none-any.whl", hash = "sha256:024ca478df22e9340661486f85298cff5f6dcdba14f3813e8830015b9ed1948f" },
Feat: Alter flask to Quart for async API serving. (#11275) ### What problem does this PR solve? #11277 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-18 17:05:16 +08:00
]
Add tavily as web searh tool. (#5349) ### What problem does this PR solve? #5198 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-02-26 10:21:04 +08:00
[[package]]
name = "tavily-python"
version = "0.5.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Add tavily as web searh tool. (#5349) ### What problem does this PR solve? #5198 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-02-26 10:21:04 +08:00
dependencies = [
{ name = "httpx" },
{ name = "requests" },
{ name = "tiktoken" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/db/ff/ba1a3769c34d022aeba544ff7b18cbcd0d23a6358fc3566b2101c6bf2817/tavily_python-0.5.1.tar.gz", hash = "sha256:44b0eefe79a057cd11d3cd03780b63b4913400122350e38285acfb502c2fffc1" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/a5/cd/71088461d7720128c78802289b3b36298f42745e5f8c334b0ffc157b881e/tavily_python-0.5.1-py3-none-any.whl", hash = "sha256:169601f703c55cf338758dcacfa7102473b479a9271d65a3af6fc3668990f757" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "tenacity"
version = "8.5.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a3/4d/6a19536c50b849338fcbe9290d562b52cbdcf30d8963d3588a68a4107df1/tenacity-8.5.0.tar.gz", hash = "sha256:8bc6c0c8a09b31e6cad13c47afbed1a567518250a9a171418582ed8d9c20ca78" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/d2/3f/8ba87d9e287b9d385a02a7114ddcef61b26f86411e121c9003eb509a1773/tenacity-8.5.0-py3-none-any.whl", hash = "sha256:b594c2a5945830c267ce6b79a166228323ed52718f30302c1359836112346687" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "tencentcloud-sdk-python"
Feat: add TCADP Parser (#10775) ### What problem does this PR solve? This PR adds a new TCADP (Tencent Cloud Advanced Document Processing) parser to RAGFlow, enabling users to leverage Tencent Cloud's document parsing capabilities for more accurate and structured document processing. The implementation includes: New TCADP Parser: A complete implementation of Tencent Cloud's document parsing API without SDK dependency Configuration Support: Added configuration options in service_conf.yaml for Tencent Cloud API credentials Frontend Integration: Updated UI components to support the new TCADP parser option Error Handling: Comprehensive error handling and retry mechanisms for API calls Result Processing: Support for both SSE streaming and JSON response formats from Tencent Cloud API ### Type of change - [x] New Feature (non-breaking change which adds functionality) --------- Co-authored-by: Kevin Hu <kevinhu.sh@gmail.com>
2025-10-27 15:14:58 +08:00
version = "3.0.1478"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "requests" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/3a/47/05163b257f6c0e60aed4272d48bdb816567ab3c805d3e8770430f0cc1be2/tencentcloud-sdk-python-3.0.1478.tar.gz", hash = "sha256:89996462d53a672946aa32d01673a4818ebcd8bc72b024f35ebe96cebe2df179" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/c5/db/daa85799b9af2aa50539b27eeb0d6a2a0ac35465f62683107847830dbe4d/tencentcloud_sdk_python-3.0.1478-py2.py3-none-any.whl", hash = "sha256:10ddee1c1348f49e2b54af606f978d4cb17fca656639e8d99b6527e6e4793833" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Update graspologic to gitee (#13362) ### What problem does this PR solve? Accelerate python module downloading ### Type of change - [x] Refactoring Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2026-03-04 17:48:47 +08:00
[[package]]
name = "text-unidecode"
version = "1.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ab/e2/e9a00f0ccb71718418230718b3d900e71a5d16e701a3dae079a21e9cd8f8/text-unidecode-1.3.tar.gz", hash = "sha256:bad6603bb14d279193107714b288be206cac565dfa49aa5b105294dd5c4aab93" }
Update graspologic to gitee (#13362) ### What problem does this PR solve? Accelerate python module downloading ### Type of change - [x] Refactoring Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2026-03-04 17:48:47 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/a6/a5/c0b6468d3824fe3fde30dbb5e1f687b291608f9473681bbf7dabbf5a87d7/text_unidecode-1.3-py2.py3-none-any.whl", hash = "sha256:1311f10e8b895935241623731c2ba64f4c455287888b18189350b67134a822e8" },
Update graspologic to gitee (#13362) ### What problem does this PR solve? Accelerate python module downloading ### Type of change - [x] Refactoring Signed-off-by: Jin Hai <haijin.chn@gmail.com>
2026-03-04 17:48:47 +08:00
]
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
[[package]]
name = "thinc"
version = "8.3.13"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
dependencies = [
{ name = "blis" },
{ name = "catalogue" },
{ name = "confection" },
{ name = "cymem" },
{ name = "murmurhash" },
{ name = "numpy" },
{ name = "packaging" },
{ name = "preshed" },
{ name = "pydantic" },
{ name = "setuptools" },
{ name = "srsly" },
{ name = "wasabi" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/13/46/76df95f2c327f9a9cef30c1523bf285627897097163584dcf5f77b2ebce2/thinc-8.3.13.tar.gz", hash = "sha256:68e658549fc1eb3ff92aed5147fcbb9c15d6e9cc0e623b4d0998d16522ffb4f9" }
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/af/b9/7b46942176df459d1804a9e77b0976f7c56f3abf3ec7485d0e5f836a0382/thinc-8.3.13-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:c2811dfd8d46d8b5d3b39051b23e64006b2994a5143b1978b436938018792af8" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a7/79/53085a72cd8f4fc4e6e313d05ea5aa98e870684f4a0fb318a9875fc0a964/thinc-8.3.13-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:5593e6300cb1ebe0c0e546e9c9fb49e7c2627a0aa688795cd4f995a8b820d2ec" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9e/3e/d61b462b16da95ac6885f95bb395e672040ee594833e571a6edcffd234f5/thinc-8.3.13-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:f697174d3fb474966ce50b430bbafa101a6d2f7ffb559dac4b5c59389ef72d22" },
{ url = "https://mirrors.aliyun.com/pypi/packages/78/4c/898cc654bb123734c71ec5a425c02ca34439517d01ce1c95a6563295580e/thinc-8.3.13-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:e9c7c5c104737b414c8c4ec578e67d78b6c859afe25cbc0684402e721415bd7f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/cd/56/1abdbf0a4ad628e8a05d6516fe0745969649d805367a3dccad8ee872981b/thinc-8.3.13-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:7a99d0e242d1ccd23f9ae6bea7cd502f8626efa65c156b91d84581d0356696c3" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f1/22/b84dbdc6be5055bbdb2a7352e2c393f67e8593c137f1b83c82bf1e062b6e/thinc-8.3.13-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:e676edd21a747afbe3e6b9f3fca8b962e36d146ded03b070cb0c28e2dfbe9499" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0f/a8/763cd7ba949334c9d2cddc92dadb68b344cb9546dc01b8d4a733dcaa16c1/thinc-8.3.13-cp313-cp313-win_amd64.whl", hash = "sha256:8ad40307f20e83f77af28ff5c6be0b86af7a8b251d1231c545508d2763157d8f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f5/15/a11f7bb3cbc97dfecf32a90552f5a8f8a5c99316a99c6c17bdabf5baf256/thinc-8.3.13-cp313-cp313-win_arm64.whl", hash = "sha256:723949cab11d1925c15447928513a718276316cec6e0de28337cca0a62be0521" },
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "threadpoolctl"
Feat: add basic Langfuse support for LLM module (#6443) ### What problem does this PR solve? #6155 Add basic Langfuse support for LLM module. A trace example: <img width="755" alt="image" src="https://github.com/user-attachments/assets/25c1f852-5116-486c-a47f-6097187142ca" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-03-24 13:18:47 +08:00
version = "3.6.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/b7/4d/08c89e34946fce2aec4fbb45c9016efd5f4d7f24af8e5d93296e935631d8/threadpoolctl-3.6.0.tar.gz", hash = "sha256:8ab8b4aa3491d812b623328249fab5302a68d2d71745c8a4c719a2fcaba9f44e" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/32/d5/f9a850d79b0851d1d4ef6456097579a9005b31fea68726a4ae5f2d82ddd9/threadpoolctl-3.6.0-py3-none-any.whl", hash = "sha256:43a0b8fd5a2928500110039e43a5eed8480b918967083ea48dc3ab9f13c4a7fb" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "thrift"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "0.22.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/b2/c2/db648cc10dd7d15560f2eafd92a27cd280811924696e0b4a87175fb28c94/thrift-0.22.0.tar.gz", hash = "sha256:42e8276afbd5f54fe1d364858b6877bc5e5a4a5ed69f6a005b94ca4918fe1466" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "tika"
version = "2.6.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "requests" },
{ name = "setuptools" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/fd/b8/055ed37d6413fef4e4af99cd7e0edc4ddfb8fc167b730b25005d212e2049/tika-2.6.0.tar.gz", hash = "sha256:56670eb812944eb25ed73f1b3b075aa41e7a135b74b240822f28b819e5b373da" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "tiktoken"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "0.12.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "regex" },
{ name = "requests" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/7d/ab/4d017d0f76ec3171d469d80fc03dfbb4e48a4bcaddaa831b31d526f05edc/tiktoken-0.12.0.tar.gz", hash = "sha256:b18ba7ee2b093863978fcb14f74b3707cdc8d4d4d3836853ce7ec60772139931" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/00/61/441588ee21e6b5cdf59d6870f86beb9789e532ee9718c251b391b70c68d6/tiktoken-0.12.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:775c2c55de2310cc1bc9a3ad8826761cbdc87770e586fd7b6da7d4589e13dab3" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1f/05/dcf94486d5c5c8d34496abe271ac76c5b785507c8eae71b3708f1ad9b45a/tiktoken-0.12.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:a01b12f69052fbe4b080a2cfb867c4de12c704b56178edf1d1d7b273561db160" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a0/70/5163fe5359b943f8db9946b62f19be2305de8c3d78a16f629d4165e2f40e/tiktoken-0.12.0-cp313-cp313-manylinux_2_28_aarch64.whl", hash = "sha256:01d99484dc93b129cd0964f9d34eee953f2737301f18b3c7257bf368d7615baa" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0c/da/c028aa0babf77315e1cef357d4d768800c5f8a6de04d0eac0f377cb619fa/tiktoken-0.12.0-cp313-cp313-manylinux_2_28_x86_64.whl", hash = "sha256:4a1a4fcd021f022bfc81904a911d3df0f6543b9e7627b51411da75ff2fe7a1be" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a0/5a/886b108b766aa53e295f7216b509be95eb7d60b166049ce2c58416b25f2a/tiktoken-0.12.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:981a81e39812d57031efdc9ec59fa32b2a5a5524d20d4776574c4b4bd2e9014a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f4/f8/4db272048397636ac7a078d22773dd2795b1becee7bc4922fe6207288d57/tiktoken-0.12.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:9baf52f84a3f42eef3ff4e754a0db79a13a27921b457ca9832cf944c6be4f8f3" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8e/32/45d02e2e0ea2be3a9ed22afc47d93741247e75018aac967b713b2941f8ea/tiktoken-0.12.0-cp313-cp313-win_amd64.whl", hash = "sha256:b8a0cd0c789a61f31bf44851defbd609e8dd1e2c8589c614cc1060940ef1f697" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ce/76/994fc868f88e016e6d05b0da5ac24582a14c47893f4474c3e9744283f1d5/tiktoken-0.12.0-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:d5f89ea5680066b68bcb797ae85219c72916c922ef0fcdd3480c7d2315ffff16" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f6/b8/57ef1456504c43a849821920d582a738a461b76a047f352f18c0b26c6516/tiktoken-0.12.0-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:b4e7ed1c6a7a8a60a3230965bdedba8cc58f68926b835e519341413370e0399a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/72/90/13da56f664286ffbae9dbcfadcc625439142675845baa62715e49b87b68b/tiktoken-0.12.0-cp313-cp313t-manylinux_2_28_aarch64.whl", hash = "sha256:fc530a28591a2d74bce821d10b418b26a094bf33839e69042a6e86ddb7a7fb27" },
{ url = "https://mirrors.aliyun.com/pypi/packages/05/df/4f80030d44682235bdaecd7346c90f67ae87ec8f3df4a3442cb53834f7e4/tiktoken-0.12.0-cp313-cp313t-manylinux_2_28_x86_64.whl", hash = "sha256:06a9f4f49884139013b138920a4c393aa6556b2f8f536345f11819389c703ebb" },
{ url = "https://mirrors.aliyun.com/pypi/packages/22/1f/ae535223a8c4ef4c0c1192e3f9b82da660be9eb66b9279e95c99288e9dab/tiktoken-0.12.0-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:04f0e6a985d95913cabc96a741c5ffec525a2c72e9df086ff17ebe35985c800e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/78/a7/f8ead382fce0243cb625c4f266e66c27f65ae65ee9e77f59ea1653b6d730/tiktoken-0.12.0-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:0ee8f9ae00c41770b5f9b0bb1235474768884ae157de3beb5439ca0fd70f3e25" },
{ url = "https://mirrors.aliyun.com/pypi/packages/93/e0/6cc82a562bc6365785a3ff0af27a2a092d57c47d7a81d9e2295d8c36f011/tiktoken-0.12.0-cp313-cp313t-win_amd64.whl", hash = "sha256:dc2dd125a62cb2b3d858484d6c614d136b5b848976794edfb63688d539b8b93f" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "tokenizers"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "0.22.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "huggingface-hub" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/73/6f/f80cfef4a312e1fb34baf7d85c72d4411afde10978d4657f8cdd811d3ccc/tokenizers-0.22.2.tar.gz", hash = "sha256:473b83b915e547aa366d1eee11806deaf419e17be16310ac0a14077f1e28f917" }
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/92/97/5dbfabf04c7e348e655e907ed27913e03db0923abb5dfdd120d7b25630e1/tokenizers-0.22.2-cp39-abi3-macosx_10_12_x86_64.whl", hash = "sha256:544dd704ae7238755d790de45ba8da072e9af3eea688f698b137915ae959281c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2e/47/174dca0502ef88b28f1c9e06b73ce33500eedfac7a7692108aec220464e7/tokenizers-0.22.2-cp39-abi3-macosx_11_0_arm64.whl", hash = "sha256:1e418a55456beedca4621dbab65a318981467a2b188e982a23e117f115ce5001" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d6/84/7990e799f1309a8b87af6b948f31edaa12a3ed22d11b352eaf4f4b2e5753/tokenizers-0.22.2-cp39-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:2249487018adec45d6e3554c71d46eb39fa8ea67156c640f7513eb26f318cec7" },
{ url = "https://mirrors.aliyun.com/pypi/packages/78/59/09d0d9ba94dcd5f4f1368d4858d24546b4bdc0231c2354aa31d6199f0399/tokenizers-0.22.2-cp39-abi3-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:25b85325d0815e86e0bac263506dd114578953b7b53d7de09a6485e4a160a7dd" },
{ url = "https://mirrors.aliyun.com/pypi/packages/47/50/b3ebb4243e7160bda8d34b731e54dd8ab8b133e50775872e7a434e524c28/tokenizers-0.22.2-cp39-abi3-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:bfb88f22a209ff7b40a576d5324bf8286b519d7358663db21d6246fb17eea2d5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e0/fa/89f4cb9e08df770b57adb96f8cbb7e22695a4cb6c2bd5f0c4f0ebcf33b66/tokenizers-0.22.2-cp39-abi3-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:1c774b1276f71e1ef716e5486f21e76333464f47bece56bbd554485982a9e03e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/64/04/ca2363f0bfbe3b3d36e95bf67e56a4c88c8e3362b658e616d1ac185d47f2/tokenizers-0.22.2-cp39-abi3-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:df6c4265b289083bf710dff49bc51ef252f9d5be33a45ee2bed151114a56207b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2e/76/932be4b50ef6ccedf9d3c6639b056a967a86258c6d9200643f01269211ca/tokenizers-0.22.2-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:369cc9fc8cc10cb24143873a0d95438bb8ee257bb80c71989e3ee290e8d72c67" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1d/28/5f9f5a4cc211b69e89420980e483831bcc29dade307955cc9dc858a40f01/tokenizers-0.22.2-cp39-abi3-musllinux_1_2_aarch64.whl", hash = "sha256:29c30b83d8dcd061078b05ae0cb94d3c710555fbb44861139f9f83dcca3dc3e4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/6c/fb/66e2da4704d6aadebf8cb39f1d6d1957df667ab24cff2326b77cda0dcb85/tokenizers-0.22.2-cp39-abi3-musllinux_1_2_armv7l.whl", hash = "sha256:37ae80a28c1d3265bb1f22464c856bd23c02a05bb211e56d0c5301a435be6c1a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/16/04/fed398b05caa87ce9b1a1bb5166645e38196081b225059a6edaff6440fac/tokenizers-0.22.2-cp39-abi3-musllinux_1_2_i686.whl", hash = "sha256:791135ee325f2336f498590eb2f11dc5c295232f288e75c99a36c5dbce63088a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/05/a1/d62dfe7376beaaf1394917e0f8e93ee5f67fea8fcf4107501db35996586b/tokenizers-0.22.2-cp39-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:38337540fbbddff8e999d59970f3c6f35a82de10053206a7562f1ea02d046fa5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fd/18/a545c4ea42af3df6effd7d13d250ba77a0a86fb20393143bbb9a92e434d4/tokenizers-0.22.2-cp39-abi3-win32.whl", hash = "sha256:a6bf3f88c554a2b653af81f3204491c818ae2ac6fbc09e76ef4773351292bc92" },
{ url = "https://mirrors.aliyun.com/pypi/packages/65/71/0670843133a43d43070abeb1949abfdef12a86d490bea9cd9e18e37c5ff7/tokenizers-0.22.2-cp39-abi3-win_amd64.whl", hash = "sha256:c9ea31edff2968b44a88f97d784c2f16dc0729b8b143ed004699ebca91f05c48" },
{ url = "https://mirrors.aliyun.com/pypi/packages/72/f4/0de46cfa12cdcbcd464cc59fde36912af405696f687e53a091fb432f694c/tokenizers-0.22.2-cp39-abi3-win_arm64.whl", hash = "sha256:9ce725d22864a1e965217204946f830c37876eee3b2ba6fc6255e8e903d5fcbc" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "tqdm"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "4.67.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "colorama", marker = "sys_platform == 'win32'" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a8/4b/29b4ef32e036bb34e4ab51796dd745cdba7ed47ad142a9f4a1eb8e0c744d/tqdm-4.67.1.tar.gz", hash = "sha256:f8aef9c52c08c13a65f30ea34f4e5aac3fd1a34959879d7e59e63027286627f2" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/d0/30/dc54f88dd4a2b5dc8a0279bdd7270e735851848b762aeb1c1184ed1f6b14/tqdm-4.67.1-py3-none-any.whl", hash = "sha256:26445eca388f82e72884e0d580d5464cd801a3ea01e63e5601bdff9ba6a48de2" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "trec-car-tools"
version = "2.6"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "cbor" },
{ name = "numpy" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/d4/71/7b62e2e56de6cdf0c648f0033a9faa41b8f712bacd71968af96277185400/trec-car-tools-2.6.tar.gz", hash = "sha256:2fce2de120224fd569b151d5bed358a4ed334e643889b9e3dfe3e5a3d15d21c8" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/36/75/661b406371f96622975eb25f9e70945d97fbe6b8e5af40342c59191962a3/trec_car_tools-2.6-py3-none-any.whl", hash = "sha256:e6f0373259e1c234222da7270ab54ca7af7a6f8d0dd32b13e158c1659d3991cf" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
[[package]]
name = "trimesh"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "4.12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
dependencies = [
{ name = "numpy" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/79/37/5cb90f04990260d2caceb6093560c6cefafca1ec522c1e43be01ca658244/trimesh-4.12.2.tar.gz", hash = "sha256:c8ca31571ac00b112e4e160e66a2d4c3491df321f056bd33806be0485d1af9d9" }
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/05/98/716a473cfb24750858ddd5d14e6527539dd206583a46408d08eeb2844a75/trimesh-4.12.2-py3-none-any.whl", hash = "sha256:b5b5afa63c5272345f2858f7676bc8c217dc8a89f4fadf6193fe10a81b5ff2aa" },
Refact: switch from google-generativeai to google-genai (#13140) ### What problem does this PR solve? Refact: switch from oogle-generativeai to google-genai #13132 Refact: commnet out unused pywencai. ### Type of change - [x] Refactoring
2026-02-24 10:28:33 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "trio"
feat: bump Python minimum from 3.12 to 3.13, drop strenum backport (#14767) Closes #14753 ## What changed | File | Change | |---|---| | `pyproject.toml` | `requires-python` → `>=3.13,<3.15`; remove `strenum==0.4.15` | | `Dockerfile` | `uv python install 3.13`, `uv sync --python 3.13` | | `.github/workflows/tests.yml` | `uv sync --python 3.13` on both matrix legs | | `CLAUDE.md` | dev setup command + requirements note updated | | `deepdoc/parser/mineru_parser.py` | `from strenum import StrEnum` → `from enum import StrEnum` | | `agent/tools/code_exec.py` | same | `StrEnum` has been in the stdlib since Python 3.11 — the `strenum` backport package is no longer needed once the floor is 3.13. ## Why uv.lock is not regenerated `uv lock --python 3.13` fails because: 1. The infiniflow/graspologic fork pins `numpy>=1.26.4,<2.0.0` 2. `tensorflow-cpu>=2.20.0` (the first release with cp313 wheels) depends on `ml-dtypes>=0.5.1`, which requires `numpy>=2.1.0` 3. These two constraints are irreconcilable on Python 3.13 The lockfile regeneration requires loosening the `numpy` upper bound in the `infiniflow/graspologic` fork. Once that fork commit is updated and the SHA in `pyproject.toml:49` is bumped, `uv lock --python 3.13` will succeed. ## RFC corrections Two claims in the original RFC (#14753) did not hold up under code review: - **"graspologic hard-blocks 3.13"** — the infiniflow fork at the pinned commit has no `<3.13` Python constraint. The blocker is the transitive `numpy<2.0.0` conflict with tensorflow-cpu's test dependency, not a direct Python version cap. - **"free-threading throughput gains for I/O-bound workload"** — Python 3.13 free-threading requires a special `--disable-gil` build and provides no benefit for async I/O code (the GIL is already released during I/O). The real motivation is forward compatibility and improved error messages.
2026-05-15 08:40:53 +02:00
version = "0.33.0"
source = { registry = "https://pypi.org/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "attrs" },
{ name = "cffi", marker = "(implementation_name != 'pypy' and os_name == 'nt' and platform_machine != 'aarch64' and sys_platform == 'linux') or (implementation_name != 'pypy' and os_name == 'nt' and sys_platform != 'darwin' and sys_platform != 'linux')" },
{ name = "idna" },
{ name = "outcome" },
{ name = "sniffio" },
{ name = "sortedcontainers" },
]
feat: bump Python minimum from 3.12 to 3.13, drop strenum backport (#14767) Closes #14753 ## What changed | File | Change | |---|---| | `pyproject.toml` | `requires-python` → `>=3.13,<3.15`; remove `strenum==0.4.15` | | `Dockerfile` | `uv python install 3.13`, `uv sync --python 3.13` | | `.github/workflows/tests.yml` | `uv sync --python 3.13` on both matrix legs | | `CLAUDE.md` | dev setup command + requirements note updated | | `deepdoc/parser/mineru_parser.py` | `from strenum import StrEnum` → `from enum import StrEnum` | | `agent/tools/code_exec.py` | same | `StrEnum` has been in the stdlib since Python 3.11 — the `strenum` backport package is no longer needed once the floor is 3.13. ## Why uv.lock is not regenerated `uv lock --python 3.13` fails because: 1. The infiniflow/graspologic fork pins `numpy>=1.26.4,<2.0.0` 2. `tensorflow-cpu>=2.20.0` (the first release with cp313 wheels) depends on `ml-dtypes>=0.5.1`, which requires `numpy>=2.1.0` 3. These two constraints are irreconcilable on Python 3.13 The lockfile regeneration requires loosening the `numpy` upper bound in the `infiniflow/graspologic` fork. Once that fork commit is updated and the SHA in `pyproject.toml:49` is bumped, `uv lock --python 3.13` will succeed. ## RFC corrections Two claims in the original RFC (#14753) did not hold up under code review: - **"graspologic hard-blocks 3.13"** — the infiniflow fork at the pinned commit has no `<3.13` Python constraint. The blocker is the transitive `numpy<2.0.0` conflict with tensorflow-cpu's test dependency, not a direct Python version cap. - **"free-threading throughput gains for I/O-bound workload"** — Python 3.13 free-threading requires a special `--disable-gil` build and provides no benefit for async I/O code (the GIL is already released during I/O). The real motivation is forward compatibility and improved error messages.
2026-05-15 08:40:53 +02:00
sdist = { url = "https://files.pythonhosted.org/packages/52/b6/c744031c6f89b18b3f5f4f7338603ab381d740a7f45938c4607b2302481f/trio-0.33.0.tar.gz", hash = "sha256:a29b92b73f09d4b48ed249acd91073281a7f1063f09caba5dc70465b5c7aa970", size = 605109, upload-time = "2026-02-14T18:40:55.386Z" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat: bump Python minimum from 3.12 to 3.13, drop strenum backport (#14767) Closes #14753 ## What changed | File | Change | |---|---| | `pyproject.toml` | `requires-python` → `>=3.13,<3.15`; remove `strenum==0.4.15` | | `Dockerfile` | `uv python install 3.13`, `uv sync --python 3.13` | | `.github/workflows/tests.yml` | `uv sync --python 3.13` on both matrix legs | | `CLAUDE.md` | dev setup command + requirements note updated | | `deepdoc/parser/mineru_parser.py` | `from strenum import StrEnum` → `from enum import StrEnum` | | `agent/tools/code_exec.py` | same | `StrEnum` has been in the stdlib since Python 3.11 — the `strenum` backport package is no longer needed once the floor is 3.13. ## Why uv.lock is not regenerated `uv lock --python 3.13` fails because: 1. The infiniflow/graspologic fork pins `numpy>=1.26.4,<2.0.0` 2. `tensorflow-cpu>=2.20.0` (the first release with cp313 wheels) depends on `ml-dtypes>=0.5.1`, which requires `numpy>=2.1.0` 3. These two constraints are irreconcilable on Python 3.13 The lockfile regeneration requires loosening the `numpy` upper bound in the `infiniflow/graspologic` fork. Once that fork commit is updated and the SHA in `pyproject.toml:49` is bumped, `uv lock --python 3.13` will succeed. ## RFC corrections Two claims in the original RFC (#14753) did not hold up under code review: - **"graspologic hard-blocks 3.13"** — the infiniflow fork at the pinned commit has no `<3.13` Python constraint. The blocker is the transitive `numpy<2.0.0` conflict with tensorflow-cpu's test dependency, not a direct Python version cap. - **"free-threading throughput gains for I/O-bound workload"** — Python 3.13 free-threading requires a special `--disable-gil` build and provides no benefit for async I/O code (the GIL is already released during I/O). The real motivation is forward compatibility and improved error messages.
2026-05-15 08:40:53 +02:00
{ url = "https://files.pythonhosted.org/packages/1c/93/dab25dc87ac48da0fe0f6419e07d0bfd98799bed4e05e7b9e0f85a1a4b4b/trio-0.33.0-py3-none-any.whl", hash = "sha256:3bd5d87f781d9b0192d592aef28691f8951d6c2e41b7e1da4c25cde6c180ae9b", size = 510294, upload-time = "2026-02-14T18:40:53.313Z" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "trio-websocket"
Add tavily as web searh tool. (#5349) ### What problem does this PR solve? #5198 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-02-26 10:21:04 +08:00
version = "0.12.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
Add tavily as web searh tool. (#5349) ### What problem does this PR solve? #5198 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-02-26 10:21:04 +08:00
{ name = "outcome" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "trio" },
{ name = "wsproto" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/d1/3c/8b4358e81f2f2cfe71b66a267f023a91db20a817b9425dd964873796980a/trio_websocket-0.12.2.tar.gz", hash = "sha256:22c72c436f3d1e264d0910a3951934798dcc5b00ae56fc4ee079d46c7cf20fae" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/c7/19/eb640a397bba49ba49ef9dbe2e7e5c04202ba045b6ce2ec36e9cadc51e04/trio_websocket-0.12.2-py3-none-any.whl", hash = "sha256:df605665f1db533f4a386c94525870851096a223adcb97f72a07e8b4beba45b6" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "typer"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "0.21.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "click" },
{ name = "rich" },
{ name = "shellingham" },
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
{ name = "typing-extensions" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/36/bf/8825b5929afd84d0dabd606c67cd57b8388cb3ec385f7ef19c5cc2202069/typer-0.21.1.tar.gz", hash = "sha256:ea835607cd752343b6b2b7ce676893e5a0324082268b48f27aa058bdb7d2145d" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/a0/1d/d9257dd49ff2ca23ea5f132edf1281a0c4f9de8a762b9ae399b670a59235/typer-0.21.1-py3-none-any.whl", hash = "sha256:7985e89081c636b88d172c2ee0cfe33c253160994d47bdfdc302defd7d1f1d01" },
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
]
[[package]]
name = "typer-slim"
version = "0.21.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
dependencies = [
{ name = "click" },
{ name = "typing-extensions" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/17/d4/064570dec6358aa9049d4708e4a10407d74c99258f8b2136bb8702303f1a/typer_slim-0.21.1.tar.gz", hash = "sha256:73495dd08c2d0940d611c5a8c04e91c2a0a98600cbd4ee19192255a233b6dbfd" }
Feat: Support multiple data sources synchronizations (#10954) ### What problem does this PR solve? #10953 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-03 19:59:18 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/c8/0a/4aca634faf693e33004796b6cee0ae2e1dba375a800c16ab8d3eff4bb800/typer_slim-0.21.1-py3-none-any.whl", hash = "sha256:6e6c31047f171ac93cc5a973c9e617dbc5ab2bddc4d0a3135dc161b4e2020e0d" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "types-requests"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "2.32.4.20260107"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "urllib3" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/0f/f3/a0663907082280664d745929205a89d41dffb29e89a50f753af7d57d0a96/types_requests-2.32.4.20260107.tar.gz", hash = "sha256:018a11ac158f801bfa84857ddec1650750e393df8a004a8a9ae2a9bec6fcb24f" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/1c/12/709ea261f2bf91ef0a26a9eed20f2623227a8ed85610c1e54c5805692ecb/types_requests-2.32.4.20260107-py3-none-any.whl", hash = "sha256:b703fe72f8ce5b31ef031264fe9395cac8f46a04661a79f7ed31a80fb308730d" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "typing-extensions"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "4.15.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/72/94/1a15dd82efb362ac84269196e94cf00f187f7ed21c242792a923cdb1c61f/typing_extensions-4.15.0.tar.gz", hash = "sha256:0cea48d173cc12fa28ecabc3b837ea3cf6f38c6d1136f85cbaaf598984861466" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/18/67/36e9267722cc04a6b9f15c7f3441c2363321a3ea07da7ae0c0707beb2a9c/typing_extensions-4.15.0-py3-none-any.whl", hash = "sha256:f0fa19c6845758ab08074a0cfa8b7aecb71c999ca73d62883bc25cc018c4e548" },
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
]
[[package]]
name = "typing-inspection"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "0.4.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
dependencies = [
{ name = "typing-extensions" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/55/e3/70399cb7dd41c10ac53367ae42139cf4b1ca5f36bb3dc6c9d33acdb43655/typing_inspection-0.4.2.tar.gz", hash = "sha256:ba561c48a67c5958007083d386c3295464928b01faa735ab8547c5692e87f464" }
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/dc/9b/47798a6c91d8bdb567fe2698fe81e0c6b7cb7ef4d13da4114b41d239f65d/typing_inspection-0.4.2-py3-none-any.whl", hash = "sha256:4ed1cacbdc298c220f1bd249ed5287caa16f34d44ef4e9c3d0cbad5b521545e7" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "tzdata"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "2025.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/5e/a7/c202b344c5ca7daf398f3b8a477eeb205cf3b6f32e7ec3a6bac0629ca975/tzdata-2025.3.tar.gz", hash = "sha256:de39c2ca5dc7b0344f2eba86f49d614019d29f060fc4ebc8a417896a620b56a7" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/c7/b0/003792df09decd6849a5e39c28b513c06e84436a54440380862b5aeff25d/tzdata-2025.3-py2.py3-none-any.whl", hash = "sha256:06a47e5700f3081aab02b2e513160914ff0694bce9947d6b76ebd6bf57cfc5d1" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
[[package]]
name = "tzlocal"
version = "5.3.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
dependencies = [
{ name = "tzdata", marker = "sys_platform == 'win32'" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/8b/2e/c14812d3d4d9cd1773c6be938f89e5735a1f11a9f184ac3639b93cef35d5/tzlocal-5.3.1.tar.gz", hash = "sha256:cceffc7edecefea1f595541dbd6e990cb1ea3d19bf01b2809f362a03dd7921fd" }
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/c2/14/e2a54fabd4f08cd7af1c07030603c3356b74da07f7cc056e600436edfa17/tzlocal-5.3.1-py3-none-any.whl", hash = "sha256:eb1a66c3ef5847adf7a834f1be0800581b683b5608e74f86ecbcef8ab91bb85d" },
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "umap-learn"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "0.5.9.post2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "numba" },
{ name = "numpy" },
{ name = "pynndescent" },
{ name = "scikit-learn" },
Bump infinity to v0.6.11. Requires python>=3.11 (#11814) ### What problem does this PR solve? Bump infinity to v0.6.11. Requires python>=3.11 ### Type of change - [x] Refactoring
2025-12-09 16:23:37 +08:00
{ name = "scipy" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "tqdm" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/5f/ee/6bc65bd375c812026a7af63fe9d09d409382120aff25f2152f1ba12af5ec/umap_learn-0.5.9.post2.tar.gz", hash = "sha256:bdf60462d779bd074ce177a0714ced17e6d161285590fa487f3f9548dd3c31c9" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/6b/b1/c24deeda9baf1fd491aaad941ed89e0fed6c583a117fd7b79e0a33a1e6c0/umap_learn-0.5.9.post2-py3-none-any.whl", hash = "sha256:fbe51166561e0e7fab00ef3d516ac2621243b8d15cf4bef9f656d701736b16a0" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "unlzw3"
version = "0.2.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/08/f1/72b313366285263aaba21a17714fbef597d7662a8737a928b2b4784eb46e/unlzw3-0.2.3.tar.gz", hash = "sha256:ede5d928c792fff9da406f20334f9739693327f448f383ae1df1774627197bbb" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/4d/fb/617af9b317ac75f5663285d3a3cc38903a76d63c6e7397768307545f4ff4/unlzw3-0.2.3-py3-none-any.whl", hash = "sha256:7760fb4f3afa1225623944c061991d89a061f7fb78665dbc4cddfdb562bb4a8b" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "uritemplate"
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
version = "4.2.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/98/60/f174043244c5306c9988380d2cb10009f91563fc4b31293d27e17201af56/uritemplate-4.2.0.tar.gz", hash = "sha256:480c2ed180878955863323eea31b0ede668795de182617fef9c6ca09e6ec9d0e" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/a9/99/3ae339466c9183ea5b8ae87b34c0b897eda475d2aec2307cae60e5cd4f29/uritemplate-4.2.0-py3-none-any.whl", hash = "sha256:962201ba1c4edcab02e60f9a0d3821e82dfc5d2d6662a21abd533879bdb8a686" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "urllib3"
Feat: support tree structured deep-research policy. (#12559) ### What problem does this PR solve? #12558 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-01-13 09:41:35 +08:00
version = "2.6.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/c7/24/5f1b3bdffd70275f6661c76461e25f024d5a38a46f04aaca912426a2b1d3/urllib3-2.6.3.tar.gz", hash = "sha256:1b62b6884944a57dbe321509ab94fd4d3b307075e0c2eae991ac71ee15ad38ed" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/39/08/aaaad47bc4e9dc8c725e68f9d04865dbcb2052843ff09c97b08904852d84/urllib3-2.6.3-py3-none-any.whl", hash = "sha256:bf272323e553dfb2e87d9bfd225ca7b0f467b919d7bbd355436d3fd37cb0acd4" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[package.optional-dependencies]
socks = [
{ name = "pysocks" },
]
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
[[package]]
name = "uuid7"
version = "0.1.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/5c/19/7472bd526591e2192926247109dbf78692e709d3e56775792fec877a7720/uuid7-0.1.0.tar.gz", hash = "sha256:8c57aa32ee7456d3cc68c95c4530bc571646defac01895cfc73545449894a63c" }
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/b5/77/8852f89a91453956582a85024d80ad96f30a41fed4c2b3dce0c9f12ecc7e/uuid7-0.1.0-py2.py3-none-any.whl", hash = "sha256:5e259bb63c8cb4aded5927ff41b444a80d0c7124e8a0ced7cf44efa1f5cccf61" },
Feat(browser control)：Add new agent component 'browser' to control browser by AI (#14888) ### What problem does this PR solve? This PR adds a new `Browser` operator to Agent workflows, enabling prompt-driven browser automation in RAGFlow.Technically based ‘Browser-Use’ It includes: - Backend browser component execution with tenant LLM integration - Upload source support (file IDs, URLs, variables, CSV/JSON array) - Downloaded file persistence to RAGFlow storage - Frontend node/operator integration, form config, icon, and i18n updates - Unit tests for upload/download and ID parsing logic - Dependency and Docker updates for browser-use runtime support ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-21 14:09:06 +08:00
]
Feat: mcp server (#7084) ### What problem does this PR solve? Add MCP support with a client example. Issue link: #4344 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-04-21 09:43:20 +08:00
[[package]]
name = "uvicorn"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "0.40.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: mcp server (#7084) ### What problem does this PR solve? Add MCP support with a client example. Issue link: #4344 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-04-21 09:43:20 +08:00
dependencies = [
{ name = "click" },
{ name = "h11" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/c3/d1/8f3c683c9561a4e6689dd3b1d345c815f10f86acd044ee1fb9a4dcd0b8c5/uvicorn-0.40.0.tar.gz", hash = "sha256:839676675e87e73694518b5574fd0f24c9d97b46bea16df7b8c05ea1a51071ea" }
Feat: mcp server (#7084) ### What problem does this PR solve? Add MCP support with a client example. Issue link: #4344 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-04-21 09:43:20 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/3d/d8/2083a1daa7439a66f3a48589a57d576aa117726762618f6bb09fe3798796/uvicorn-0.40.0-py3-none-any.whl", hash = "sha256:c6c8f55bc8bf13eb6fa9ff87ad62308bbbc33d0b67f84293151efe87e0d5f2ee" },
Feat: mcp server (#7084) ### What problem does this PR solve? Add MCP support with a client example. Issue link: #4344 ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-04-21 09:43:20 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "valkey"
version = "6.0.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/45/f7/b552b7a67017e6233cd8a3b783ce8c4b548e29df98daedd7fb4c4c2cc8f8/valkey-6.0.2.tar.gz", hash = "sha256:dc2e91512b82d1da0b91ab0cdbd8c97c0c0250281728cb32f9398760df9caeae" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/d7/cb/b1eac0fe9cbdbba0a5cf189f5778fe54ba7d7c9f26c2f62ca8d759b38f52/valkey-6.0.2-py3-none-any.whl", hash = "sha256:dbbdd65439ee0dc5689502c54f1899504cc7268e85cb7fe8935f062178ff5805" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "volcengine"
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
version = "1.0.194"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "google" },
{ name = "protobuf" },
{ name = "pycryptodome" },
{ name = "pytz" },
{ name = "requests" },
{ name = "retry" },
{ name = "six" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/21/6d/0b29d9bb3895990391ec1e3722f153c24f94a4f1bea2d2d4f418050fae89/volcengine-1.0.194.tar.gz", hash = "sha256:cab0ea38291ca7b2bbffe130a7b173cf6fdc4a1af61cf7792c35296d5498766c" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "voyageai"
version = "0.2.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "aiohttp" },
{ name = "aiolimiter" },
{ name = "numpy" },
{ name = "requests" },
{ name = "tenacity" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/7e/8c/23240073e306e6f49f6d2a33de28ca74fe36ebcd34bca3cfbcedcdd0ce63/voyageai-0.2.3.tar.gz", hash = "sha256:28322aa7a64cdaa774be6fcf3e4fd6a08694ea25acd5fadd1eff1b8ef8dab68a" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/bf/7c/43fb4689fe287eceb701f389863aab35211835d63bbb9a798cfefa80d7de/voyageai-0.2.3-py3-none-any.whl", hash = "sha256:59c4958bd991e83cedb5a82d5e14ac698ce67e42713ea10467631a48ee272b15" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "warc3-wet"
version = "0.2.5"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/21/c6/24c9b4a2b2b1741b57d7f44ff9790eb4ef28de898c17c2b1ca1efabf8c96/warc3_wet-0.2.5.tar.gz", hash = "sha256:15e50402dabaa1e95307f1e2a6169cfd5f137b70761d9f0b16a10aa6de227970" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/f4/99/0a5582a106679fd9439af51631b6c6cb627fd96cbc85a02927e6812605b8/warc3_wet-0.2.5-py3-none-any.whl", hash = "sha256:5a9a525383fb1af159734baa75f349a7c4ec7bccd1b938681b5748515d2bf624" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "warc3-wet-clueweb09"
version = "0.2.5"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/9f/c1/dd817bf57e0274dacb10e0ac868cb6cd70876950cf361c41879c030a2b8b/warc3-wet-clueweb09-0.2.5.tar.gz", hash = "sha256:3054bfc07da525d5967df8ca3175f78fa3f78514c82643f8c81fbca96300b836" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
[[package]]
name = "wasabi"
version = "1.1.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
dependencies = [
{ name = "colorama", marker = "sys_platform == 'win32'" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ac/f9/054e6e2f1071e963b5e746b48d1e3727470b2a490834d18ad92364929db3/wasabi-1.1.3.tar.gz", hash = "sha256:4bb3008f003809db0c3e28b4daf20906ea871a2bb43f9914197d540f4f2e0878" }
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/06/7c/34330a89da55610daa5f245ddce5aab81244321101614751e7537f125133/wasabi-1.1.3-py3-none-any.whl", hash = "sha256:f76e16e8f7e79f8c4c8be49b4024ac725713ab10cd7f19350ad18a8e3f71728c" },
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "wcwidth"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "0.2.14"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/24/30/6b0809f4510673dc723187aeaf24c7f5459922d01e2f794277a3dfb90345/wcwidth-0.2.14.tar.gz", hash = "sha256:4d478375d31bc5395a3c55c40ccdf3354688364cd61c4f6adacaa9215d0b3605" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/af/b5/123f13c975e9f27ab9c0770f514345bd406d0e8d3b7a0723af9d43f710af/wcwidth-0.2.14-py2.py3-none-any.whl", hash = "sha256:a7bb560c8aee30f9957e5f9895805edd20602f2d7f720186dfd906e82b4982e1" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
[[package]]
name = "weasel"
version = "1.0.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
dependencies = [
{ name = "cloudpathlib" },
{ name = "confection" },
{ name = "httpx" },
{ name = "packaging" },
{ name = "pydantic" },
{ name = "smart-open" },
{ name = "srsly" },
{ name = "typer" },
{ name = "wasabi" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/ce/e5/e272bb9a045105a1fdf4b798d8086f5932a178f4d738f17a74f5c9e0ae9a/weasel-1.0.0.tar.gz", hash = "sha256:7b129b44c90cc543b760532974ca1e4eb30dad2aa2026f57bdce66354ae610fc" }
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/0a/07/57ebf7a6798b016c064bd0ca81b4c6a99daa4dc377b898bc7b41eb6b5af0/weasel-1.0.0-py3-none-any.whl", hash = "sha256:89518acee027f49d743126c3502d35e6dd14f5768be5c37c9af47c171b6005cc" },
GraphRAG feature - Part 1 - add spacy to extract entity and relation (#14670) ### What problem does this PR solve? GraphRAG feature - Part 1 - add spacy to extract entity and relation <img width="1621" height="1288" alt="image" src="https://github.com/user-attachments/assets/aadeddad-94da-46c6-adad-9c3784181f61" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2026-05-11 12:59:59 +08:00
]
Feat: Add Webdav storage as data source (#11422) ### What problem does this PR solve? This PR adds webdav storage as data source for data sync service. ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-26 07:14:42 +01:00
[[package]]
name = "webdav4"
version = "0.10.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: Add Webdav storage as data source (#11422) ### What problem does this PR solve? This PR adds webdav storage as data source for data sync service. ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-26 07:14:42 +01:00
dependencies = [
{ name = "httpx" },
{ name = "python-dateutil" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/08/3d/d604f9d5195689e578f124f196a5d7e80f3106c8404f5c19b2181691de19/webdav4-0.10.0.tar.gz", hash = "sha256:387da6f0ee384e77149dddd9bcfd434afa155882f6c440a529a7cb458624407f" }
Feat: Add Webdav storage as data source (#11422) ### What problem does this PR solve? This PR adds webdav storage as data source for data sync service. ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-26 07:14:42 +01:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/60/02/1b77232297fa52f7bedcf70f3ebe3817e9295f302389fb57dd0e6c077329/webdav4-0.10.0-py3-none-any.whl", hash = "sha256:8f915d72483e572089a3af0a2ad20c7e12d04eee9b9134eb718dbfa37af221d8" },
Feat: Add Webdav storage as data source (#11422) ### What problem does this PR solve? This PR adds webdav storage as data source for data sync service. ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-11-26 07:14:42 +01:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "webdriver-manager"
version = "4.0.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "packaging" },
{ name = "python-dotenv" },
{ name = "requests" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/e5/50/2958aa25647e86334b30b4f8c819cc4fd5f15d3d0115042a4c924ec6e94d/webdriver_manager-4.0.1.tar.gz", hash = "sha256:25ec177c6a2ce9c02fb8046f1b2732701a9418d6a977967bb065d840a3175d87" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/b1/51/b5c11cf739ac4eecde611794a0ec9df420d0239d51e73bc19eb44f02b48b/webdriver_manager-4.0.1-py2.py3-none-any.whl", hash = "sha256:d7970052295bb9cda2c1a24cf0b872dd2c41ababcc78f7b6b8dc37a41e979a7e" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "webencodings"
version = "0.5.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/0b/02/ae6ceac1baeda530866a85075641cec12989bd8d31af6d5ab4a3e8c92f47/webencodings-0.5.1.tar.gz", hash = "sha256:b36a1c245f2d304965eb4e0a82848379241dc04b865afcc4aab16748587e1923" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/f4/24/2a3e3df732393fed8b3ebf2ec078f05546de641fe1b667ee316ec1dcf3b7/webencodings-0.5.1-py2.py3-none-any.whl", hash = "sha256:a0af1213f3c2226497a97e2b3aa01a7e4bee4f403f95be16fc9acd2947514a78" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "websocket-client"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "1.9.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/2c/41/aa4bf9664e4cda14c3b39865b12251e8e7d239f4cd0e3cc1b6c2ccde25c1/websocket_client-1.9.0.tar.gz", hash = "sha256:9e813624b6eb619999a97dc7958469217c3176312b3a16a4bd1bc7e08a46ec98" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/34/db/b10e48aa8fff7407e67470363eac595018441cf32d5e1001567a7aeba5d2/websocket_client-1.9.0-py3-none-any.whl", hash = "sha256:af248a825037ef591efbf6ed20cc5faa03d3b47b9e5a2230a529eeee1c1fc3ef" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
[[package]]
name = "websockets"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "15.0.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/21/e6/26d09fab466b7ca9c7737474c52be4f76a40301b08362eb2dbc19dcc16c1/websockets-15.0.1.tar.gz", hash = "sha256:82544de02076bafba038ce055ee6412d68da13ab47f0c60cab827346de828dee" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/cb/9f/51f0cf64471a9d2b4d0fc6c534f323b664e7095640c34562f5182e5a7195/websockets-15.0.1-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:ee443ef070bb3b6ed74514f5efaa37a252af57c90eb33b956d35c8e9c10a1931" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8a/05/aa116ec9943c718905997412c5989f7ed671bc0188ee2ba89520e8765d7b/websockets-15.0.1-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:5a939de6b7b4e18ca683218320fc67ea886038265fd1ed30173f5ce3f8e85675" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ff/0b/33cef55ff24f2d92924923c99926dcce78e7bd922d649467f0eda8368923/websockets-15.0.1-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:746ee8dba912cd6fc889a8147168991d50ed70447bf18bcda7039f7d2e3d9151" },
{ url = "https://mirrors.aliyun.com/pypi/packages/31/1d/063b25dcc01faa8fada1469bdf769de3768b7044eac9d41f734fd7b6ad6d/websockets-15.0.1-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:595b6c3969023ecf9041b2936ac3827e4623bfa3ccf007575f04c5a6aa318c22" },
{ url = "https://mirrors.aliyun.com/pypi/packages/93/53/9a87ee494a51bf63e4ec9241c1ccc4f7c2f45fff85d5bde2ff74fcb68b9e/websockets-15.0.1-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:3c714d2fc58b5ca3e285461a4cc0c9a66bd0e24c5da9911e30158286c9b5be7f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ff/b2/83a6ddf56cdcbad4e3d841fcc55d6ba7d19aeb89c50f24dd7e859ec0805f/websockets-15.0.1-cp313-cp313-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0f3c1e2ab208db911594ae5b4f79addeb3501604a165019dd221c0bdcabe4db8" },
{ url = "https://mirrors.aliyun.com/pypi/packages/98/41/e7038944ed0abf34c45aa4635ba28136f06052e08fc2168520bb8b25149f/websockets-15.0.1-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:229cf1d3ca6c1804400b0a9790dc66528e08a6a1feec0d5040e8b9eb14422375" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e0/17/de15b6158680c7623c6ef0db361da965ab25d813ae54fcfeae2e5b9ef910/websockets-15.0.1-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:756c56e867a90fb00177d530dca4b097dd753cde348448a1012ed6c5131f8b7d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/33/2b/1f168cb6041853eef0362fb9554c3824367c5560cbdaad89ac40f8c2edfc/websockets-15.0.1-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:558d023b3df0bffe50a04e710bc87742de35060580a293c2a984299ed83bc4e4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/86/eb/20b6cdf273913d0ad05a6a14aed4b9a85591c18a987a3d47f20fa13dcc47/websockets-15.0.1-cp313-cp313-win32.whl", hash = "sha256:ba9e56e8ceeeedb2e080147ba85ffcd5cd0711b89576b83784d8605a7df455fa" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1b/6c/c65773d6cab416a64d191d6ee8a8b1c68a09970ea6909d16965d26bfed1e/websockets-15.0.1-cp313-cp313-win_amd64.whl", hash = "sha256:e09473f095a819042ecb2ab9465aee615bd9c2028e4ef7d933600a8401c79561" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fa/a8/5b41e0da817d64113292ab1f8247140aac61cbf6cfd085d6a0fa77f4984f/websockets-15.0.1-py3-none-any.whl", hash = "sha256:f7a866fbc1e97b5c617ee4116daaa09b722101d4a3c170c787450ba409f9736f" },
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
]
[[package]]
name = "wechatpy"
version = "1.8.18"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
dependencies = [
{ name = "optionaldict" },
{ name = "python-dateutil" },
{ name = "requests" },
{ name = "six" },
{ name = "xmltodict" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/80/34/849aaa2a03dc41b002e224006746bdb2eea729435225709139c4b3a95d2b/wechatpy-1.8.18.tar.gz", hash = "sha256:b356d0181abcf8487268e5759f942c4ffbb5090c00197858109e90000a301d08" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/09/d6/a49a4571affd12dbb88067d7c3a874dc069ffa471c29fd3a8e2771e7f0c2/wechatpy-1.8.18-py2.py3-none-any.whl", hash = "sha256:bfd0457dfd3f4623a490534cc9cb9c29b53f9397b3ebbee0f11ee7925b9b71f6" },
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
]
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "werkzeug"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "3.1.5"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "markupsafe" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/5a/70/1469ef1d3542ae7c2c7b72bd5e3a4e6ee69d7978fa8a3af05a38eca5becf/werkzeug-3.1.5.tar.gz", hash = "sha256:6a548b0e88955dd07ccb25539d7d0cc97417ee9e179677d22c7041c8f078ce67" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/ad/e4/8d97cca767bcc1be76d16fb76951608305561c6e056811587f36cb1316a8/werkzeug-3.1.5-py3-none-any.whl", hash = "sha256:5111e36e91086ece91f93268bb39b4a35c1e6f1feac762c9c822ded0a4e322dc" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "wikipedia"
version = "1.4.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "beautifulsoup4" },
{ name = "requests" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/67/35/25e68fbc99e672127cc6fbb14b8ec1ba3dfef035bf1e4c90f78f24a80b7d/wikipedia-1.4.0.tar.gz", hash = "sha256:db0fad1829fdd441b1852306e9856398204dc0786d2996dd2e0c8bb8e26133b2" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
[[package]]
name = "win-unicode-console"
version = "0.5"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/89/8d/7aad74930380c8972ab282304a2ff45f3d4927108bb6693cabcc9fc6a099/win_unicode_console-0.5.zip", hash = "sha256:d4142d4d56d46f449d6f00536a73625a871cba040f0bc1a2e305a04578f07d1e" }
Feat: Pipeline Docx file supports Markdown output (#10439) ### What problem does this PR solve? Pipeline Docx file supports Markdown output. <img width="1242" height="755" alt="image" src="https://github.com/user-attachments/assets/63cca75b-20b9-4a90-a01c-c0c2fccf1f2a" /> <img width="1227" height="717" alt="image" src="https://github.com/user-attachments/assets/0dcb94b2-7ba0-48d5-9231-dc6e5c4b4192" /> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-10-10 09:39:15 +08:00
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "word2number"
version = "1.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/4a/29/a31940c848521f0725f0df6b25dca8917f13a2025b0e8fcbe5d0457e45e6/word2number-1.1.zip", hash = "sha256:70e27a5d387f67b04c71fbb7621c05930b19bfd26efd6851e6e0f9969dcde7d0" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
[[package]]
name = "wrapt"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "1.17.3"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/95/8f/aeb76c5b46e273670962298c23e7ddde79916cb74db802131d49a85e4b7d/wrapt-1.17.3.tar.gz", hash = "sha256:f66eb08feaa410fe4eebd17f2a2c8e2e46d3476e9f8c783daa8e09e0faa666d0" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/fc/f6/759ece88472157acb55fc195e5b116e06730f1b651b5b314c66291729193/wrapt-1.17.3-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:a47681378a0439215912ef542c45a783484d4dd82bac412b71e59cf9c0e1cea0" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4f/a9/49940b9dc6d47027dc850c116d79b4155f15c08547d04db0f07121499347/wrapt-1.17.3-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:54a30837587c6ee3cd1a4d1c2ec5d24e77984d44e2f34547e2323ddb4e22eb77" },
{ url = "https://mirrors.aliyun.com/pypi/packages/45/35/6a08de0f2c96dcdd7fe464d7420ddb9a7655a6561150e5fc4da9356aeaab/wrapt-1.17.3-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:16ecf15d6af39246fe33e507105d67e4b81d8f8d2c6598ff7e3ca1b8a37213f7" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0c/37/6faf15cfa41bf1f3dba80cd3f5ccc6622dfccb660ab26ed79f0178c7497f/wrapt-1.17.3-cp313-cp313-manylinux1_x86_64.manylinux_2_28_x86_64.manylinux_2_5_x86_64.whl", hash = "sha256:6fd1ad24dc235e4ab88cda009e19bf347aabb975e44fd5c2fb22a3f6e4141277" },
{ url = "https://mirrors.aliyun.com/pypi/packages/78/f2/efe19ada4a38e4e15b6dff39c3e3f3f73f5decf901f66e6f72fe79623a06/wrapt-1.17.3-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:0ed61b7c2d49cee3c027372df5809a59d60cf1b6c2f81ee980a091f3afed6a2d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/40/90/ca86701e9de1622b16e09689fc24b76f69b06bb0150990f6f4e8b0eeb576/wrapt-1.17.3-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:423ed5420ad5f5529db9ce89eac09c8a2f97da18eb1c870237e84c5a5c2d60aa" },
{ url = "https://mirrors.aliyun.com/pypi/packages/fd/e0/d10bd257c9a3e15cbf5523025252cc14d77468e8ed644aafb2d6f54cb95d/wrapt-1.17.3-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:e01375f275f010fcbf7f643b4279896d04e571889b8a5b3f848423d91bf07050" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e8/cf/7d848740203c7b4b27eb55dbfede11aca974a51c3d894f6cc4b865f42f58/wrapt-1.17.3-cp313-cp313-win32.whl", hash = "sha256:53e5e39ff71b3fc484df8a522c933ea2b7cdd0d5d15ae82e5b23fde87d44cbd8" },
{ url = "https://mirrors.aliyun.com/pypi/packages/57/54/35a84d0a4d23ea675994104e667ceff49227ce473ba6a59ba2c84f250b74/wrapt-1.17.3-cp313-cp313-win_amd64.whl", hash = "sha256:1f0b2f40cf341ee8cc1a97d51ff50dddb9fcc73241b9143ec74b30fc4f44f6cb" },
{ url = "https://mirrors.aliyun.com/pypi/packages/01/77/66e54407c59d7b02a3c4e0af3783168fff8e5d61def52cda8728439d86bc/wrapt-1.17.3-cp313-cp313-win_arm64.whl", hash = "sha256:7425ac3c54430f5fc5e7b6f41d41e704db073309acfc09305816bc6a0b26bb16" },
{ url = "https://mirrors.aliyun.com/pypi/packages/1f/f6/a933bd70f98e9cf3e08167fc5cd7aaaca49147e48411c0bd5ae701bb2194/wrapt-1.17.3-py3-none-any.whl", hash = "sha256:7171ae35d2c33d326ac19dd8facb1e82e5fd04ef8c6c0e394d7af55a55051c22" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "wsproto"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "1.3.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "h11" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/c7/79/12135bdf8b9c9367b8701c2c19a14c913c120b882d50b014ca0d38083c2c/wsproto-1.3.2.tar.gz", hash = "sha256:b86885dcf294e15204919950f666e06ffc6c7c114ca900b060d6e16293528294" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/a4/f5/10b68b7b1544245097b2a1b8238f66f2fc6dcaeb24ba5d917f52bd2eed4f/wsproto-1.3.2-py3-none-any.whl", hash = "sha256:61eea322cdf56e8cc904bd3ad7573359a242ba65688716b0710a5eb12beab584" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "xgboost"
Fix: revert xgboost version to 1.6.0 (#13984) ### What problem does this PR solve? Revert xgboost version to 1.6.0 ### Type of change - [ ] Bug Fix (non-breaking change which fixes an issue) - [ ] New Feature (non-breaking change which adds functionality) - [ ] Documentation Update - [ ] Refactoring - [ ] Performance Improvement - [ ] Other (please describe): <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **Chores** * Updated xgboost dependency from version 3.2.0 to 1.6.0 <!-- end of auto-generated comment: release notes by coderabbit.ai -->
2026-04-08 19:53:47 +08:00
version = "1.6.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "numpy" },
Bump infinity to v0.6.11. Requires python>=3.11 (#11814) ### What problem does this PR solve? Bump infinity to v0.6.11. Requires python>=3.11 ### Type of change - [x] Refactoring
2025-12-09 16:23:37 +08:00
{ name = "scipy" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/77/89/92b399140a7688443fc182b54240822c903e906121d63446eb2f84350e99/xgboost-1.6.0.tar.gz", hash = "sha256:9c944c2495cb426b8a365021565755c39ee0b53156cf5e53a4346bdad2e3b734" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/f1/71/abca2240b5d19aa3e90c8228cf307962fc9f598acc3c623fb49db83b4092/xgboost-1.6.0-py3-none-macosx_10_15_x86_64.macosx_11_0_x86_64.macosx_12_0_x86_64.whl", hash = "sha256:5f7fd61024c41d0c424a8272dfd27797a0393a616b717c05c0f981a49a47b4fd" },
{ url = "https://mirrors.aliyun.com/pypi/packages/49/d0/85c9c40e7ca1a4bc05278c1e57a89c43ab846be4cb5227871ca7605921a6/xgboost-1.6.0-py3-none-macosx_12_0_arm64.whl", hash = "sha256:ad27c6a72f6abef6d20e67f957fb25553bb09a6d1c4eaf08cb8ee7efca288255" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c3/be/18970943eb7e9d9ded5e37e87c1dc02c8a961416f725f2734629f26d69d5/xgboost-1.6.0-py3-none-manylinux2014_aarch64.whl", hash = "sha256:b1d532b8d548dd3acb4bd5f56632339e48167d9e2ec0eda2d8d6b4cd772e03b4" },
{ url = "https://mirrors.aliyun.com/pypi/packages/bf/64/c467a20848adc3d1c3f45d60df9c7cd0c40a548fd534a9f842a35114039d/xgboost-1.6.0-py3-none-manylinux2014_x86_64.whl", hash = "sha256:640b9649104f22f0dc43c7202d22cde5531cc303801a9c75cad3f2b6e413dcf7" },
{ url = "https://mirrors.aliyun.com/pypi/packages/64/51/3e33a4df0ca66474e7f4e357328a5c7b35fb52cbc48b312c64d276d37da8/xgboost-1.6.0-py3-none-win_amd64.whl", hash = "sha256:e2f9baca0b7cbc208ad4fbafa4cd70b50b292717ee8ba817a3ba7a0fe49de958" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "xlrd"
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
version = "2.0.2"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/07/5a/377161c2d3538d1990d7af382c79f3b2372e880b65de21b01b1a2b78691e/xlrd-2.0.2.tar.gz", hash = "sha256:08b5e25de58f21ce71dc7db3b3b8106c1fa776f3024c54e45b45b374e89234c9" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/1a/62/c8d562e7766786ba6587d09c5a8ba9f718ed3fa8af7f4553e8f91c36f302/xlrd-2.0.2-py2.py3-none-any.whl", hash = "sha256:ea762c3d29f4cca48d82df517b6d89fbce4db3107f9d78713e48cd321d5c9aa9" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "xlsxwriter"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "3.2.9"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/46/2c/c06ef49dc36e7954e55b802a8b231770d286a9758b3d936bd1e04ce5ba88/xlsxwriter-3.2.9.tar.gz", hash = "sha256:254b1c37a368c444eac6e2f867405cc9e461b0ed97a3233b2ac1e574efb4140c" }
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/3a/0c/3662f4a66880196a590b202f0db82d919dd2f89e99a27fadef91c4a33d41/xlsxwriter-3.2.9-py3-none-any.whl", hash = "sha256:9a5db42bc5dff014806c58a20b9eae7322a134abb6fce3c92c181bfb275ec5b3" },
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
]
[[package]]
name = "xmltodict"
version = "1.0.4"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/19/70/80f3b7c10d2630aa66414bf23d210386700aa390547278c789afa994fd7e/xmltodict-1.0.4.tar.gz", hash = "sha256:6d94c9f834dd9e44514162799d344d815a3a4faec913717a9ecbfa5be1bb8e61" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/38/34/98a2f52245f4d47be93b580dae5f9861ef58977d73a79eb47c58f1ad1f3a/xmltodict-1.0.4-py3-none-any.whl", hash = "sha256:a4a00d300b0e1c59fc2bfccb53d7b2e88c32f200df138a0dd2229f842497026a" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "xpinyin"
version = "0.7.6"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/16/f2/d548d2f91106644b1b51df4cc59c1b3fabe9048954f18011775250c32d53/xpinyin-0.7.6.tar.gz", hash = "sha256:dec6aa0f4d9f9b6788d8131245293f1951180333a6d474b467b2d556221862fe" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/a5/30/40e099a8da32105c8adf996abe92a5bbf5ecd338de2c4cc491b5718299ce/xpinyin-0.7.6-py3-none-any.whl", hash = "sha256:1d78eac9f612c20e155d7c3eb9dd7f9d3ec4e2667c52049e990b8bd036171a52" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "xxhash"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "3.6.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/02/84/30869e01909fb37a6cc7e18688ee8bf1e42d57e7e0777636bd47524c43c7/xxhash-3.6.0.tar.gz", hash = "sha256:f0162a78b13a0d7617b2845b90c763339d1f1d82bb04a4b07f4ab535cc5e05d6" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/33/76/35d05267ac82f53ae9b0e554da7c5e281ee61f3cad44c743f0fcd354f211/xxhash-3.6.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:599e64ba7f67472481ceb6ee80fa3bd828fd61ba59fb11475572cc5ee52b89ec" },
{ url = "https://mirrors.aliyun.com/pypi/packages/31/a8/3fbce1cd96534a95e35d5120637bf29b0d7f5d8fa2f6374e31b4156dd419/xxhash-3.6.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:7d8b8aaa30fca4f16f0c84a5c8d7ddee0e25250ec2796c973775373257dde8f1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0c/ea/d387530ca7ecfa183cb358027f1833297c6ac6098223fd14f9782cd0015c/xxhash-3.6.0-cp313-cp313-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:d597acf8506d6e7101a4a44a5e428977a51c0fadbbfd3c39650cca9253f6e5a6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ba/0c/71435dcb99874b09a43b8d7c54071e600a7481e42b3e3ce1eb5226a5711a/xxhash-3.6.0-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:858dc935963a33bc33490128edc1c12b0c14d9c7ebaa4e387a7869ecc4f3e263" },
{ url = "https://mirrors.aliyun.com/pypi/packages/84/7a/c2b3d071e4bb4a90b7057228a99b10d51744878f4a8a6dd643c8bd897620/xxhash-3.6.0-cp313-cp313-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:ba284920194615cb8edf73bf52236ce2e1664ccd4a38fdb543506413529cc546" },
{ url = "https://mirrors.aliyun.com/pypi/packages/81/5f/640b6eac0128e215f177df99eadcd0f1b7c42c274ab6a394a05059694c5a/xxhash-3.6.0-cp313-cp313-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:4b54219177f6c6674d5378bd862c6aedf64725f70dd29c472eaae154df1a2e89" },
{ url = "https://mirrors.aliyun.com/pypi/packages/5e/1e/3c3d3ef071b051cc3abbe3721ffb8365033a172613c04af2da89d5548a87/xxhash-3.6.0-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:42c36dd7dbad2f5238950c377fcbf6811b1cdb1c444fab447960030cea60504d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2c/bd/4a5f68381939219abfe1c22a9e3a5854a4f6f6f3c4983a87d255f21f2e5d/xxhash-3.6.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:f22927652cba98c44639ffdc7aaf35828dccf679b10b31c4ad72a5b530a18eb7" },
{ url = "https://mirrors.aliyun.com/pypi/packages/eb/37/b80fe3d5cfb9faff01a02121a0f4d565eb7237e9e5fc66e73017e74dcd36/xxhash-3.6.0-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:b45fad44d9c5c119e9c6fbf2e1c656a46dc68e280275007bbfd3d572b21426db" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d7/fd/2c0a00c97b9e18f72e1f240ad4e8f8a90fd9d408289ba9c7c495ed7dc05c/xxhash-3.6.0-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:6f2580ffab1a8b68ef2b901cde7e55fa8da5e4be0977c68f78fc80f3c143de42" },
{ url = "https://mirrors.aliyun.com/pypi/packages/93/86/5dd8076a926b9a95db3206aba20d89a7fc14dd5aac16e5c4de4b56033140/xxhash-3.6.0-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:40c391dd3cd041ebc3ffe6f2c862f402e306eb571422e0aa918d8070ba31da11" },
{ url = "https://mirrors.aliyun.com/pypi/packages/af/3c/0bb129170ee8f3650f08e993baee550a09593462a5cddd8e44d0011102b1/xxhash-3.6.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:f205badabde7aafd1a31e8ca2a3e5a763107a71c397c4481d6a804eb5063d8bd" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e9/3a/6797e0114c21d1725e2577508e24006fd7ff1d8c0c502d3b52e45c1771d8/xxhash-3.6.0-cp313-cp313-win32.whl", hash = "sha256:2577b276e060b73b73a53042ea5bd5203d3e6347ce0d09f98500f418a9fcf799" },
{ url = "https://mirrors.aliyun.com/pypi/packages/86/15/9bc32671e9a38b413a76d24722a2bf8784a132c043063a8f5152d390b0f9/xxhash-3.6.0-cp313-cp313-win_amd64.whl", hash = "sha256:757320d45d2fbcce8f30c42a6b2f47862967aea7bf458b9625b4bbe7ee390392" },
{ url = "https://mirrors.aliyun.com/pypi/packages/39/c5/cc01e4f6188656e56112d6a8e0dfe298a16934b8c47a247236549a3f7695/xxhash-3.6.0-cp313-cp313-win_arm64.whl", hash = "sha256:457b8f85dec5825eed7b69c11ae86834a018b8e3df5e77783c999663da2f96d6" },
{ url = "https://mirrors.aliyun.com/pypi/packages/f3/30/25e5321c8732759e930c555176d37e24ab84365482d257c3b16362235212/xxhash-3.6.0-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:a42e633d75cdad6d625434e3468126c73f13f7584545a9cf34e883aa1710e702" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9f/3c/0573299560d7d9f8ab1838f1efc021a280b5ae5ae2e849034ef3dee18810/xxhash-3.6.0-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:568a6d743219e717b07b4e03b0a828ce593833e498c3b64752e0f5df6bfe84db" },
{ url = "https://mirrors.aliyun.com/pypi/packages/7a/1c/52d83a06e417cd9d4137722693424885cc9878249beb3a7c829e74bf7ce9/xxhash-3.6.0-cp313-cp313t-manylinux1_i686.manylinux_2_28_i686.manylinux_2_5_i686.whl", hash = "sha256:bec91b562d8012dae276af8025a55811b875baace6af510412a5e58e3121bc54" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e3/8e/c6d158d12a79bbd0b878f8355432075fc82759e356ab5a111463422a239b/xxhash-3.6.0-cp313-cp313t-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:78e7f2f4c521c30ad5e786fdd6bae89d47a32672a80195467b5de0480aa97b1f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/bc/68/c4c80614716345d55071a396cf03d06e34b5f4917a467faf43083c995155/xxhash-3.6.0-cp313-cp313t-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:3ed0df1b11a79856df5ffcab572cbd6b9627034c1c748c5566fa79df9048a7c5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/7e/e9/ae27c8ffec8b953efa84c7c4a6c6802c263d587b9fc0d6e7cea64e08c3af/xxhash-3.6.0-cp313-cp313t-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:0e4edbfc7d420925b0dd5e792478ed393d6e75ff8fc219a6546fb446b6a417b1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d7/6b/33e21afb1b5b3f46b74b6bd1913639066af218d704cc0941404ca717fc57/xxhash-3.6.0-cp313-cp313t-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:fba27a198363a7ef87f8c0f6b171ec36b674fe9053742c58dd7e3201c1ab30ee" },
{ url = "https://mirrors.aliyun.com/pypi/packages/96/b6/fcabd337bc5fa624e7203aa0fa7d0c49eed22f72e93229431752bddc83d9/xxhash-3.6.0-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:794fe9145fe60191c6532fa95063765529770edcdd67b3d537793e8004cabbfd" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4b/d3/9ee6160e644d660fcf176c5825e61411c7f62648728f69c79ba237250143/xxhash-3.6.0-cp313-cp313t-musllinux_1_2_i686.whl", hash = "sha256:6105ef7e62b5ac73a837778efc331a591d8442f8ef5c7e102376506cb4ae2729" },
{ url = "https://mirrors.aliyun.com/pypi/packages/0d/98/e8de5baa5109394baf5118f5e72ab21a86387c4f89b0e77ef3e2f6b0327b/xxhash-3.6.0-cp313-cp313t-musllinux_1_2_ppc64le.whl", hash = "sha256:f01375c0e55395b814a679b3eea205db7919ac2af213f4a6682e01220e5fe292" },
{ url = "https://mirrors.aliyun.com/pypi/packages/7b/1d/71056535dec5c3177eeb53e38e3d367dd1d16e024e63b1cee208d572a033/xxhash-3.6.0-cp313-cp313t-musllinux_1_2_s390x.whl", hash = "sha256:d706dca2d24d834a4661619dcacf51a75c16d65985718d6a7d73c1eeeb903ddf" },
{ url = "https://mirrors.aliyun.com/pypi/packages/dc/6c/5cbde9de2cd967c322e651c65c543700b19e7ae3e0aae8ece3469bf9683d/xxhash-3.6.0-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:5f059d9faeacd49c0215d66f4056e1326c80503f51a1532ca336a385edadd033" },
{ url = "https://mirrors.aliyun.com/pypi/packages/19/fa/0172e350361d61febcea941b0cc541d6e6c8d65d153e85f850a7b256ff8a/xxhash-3.6.0-cp313-cp313t-win32.whl", hash = "sha256:1244460adc3a9be84731d72b8e80625788e5815b68da3da8b83f78115a40a7ec" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ad/e6/e8cf858a2b19d6d45820f072eff1bea413910592ff17157cabc5f1227a16/xxhash-3.6.0-cp313-cp313t-win_amd64.whl", hash = "sha256:b1e420ef35c503869c4064f4a2f2b08ad6431ab7b229a05cce39d74268bca6b8" },
{ url = "https://mirrors.aliyun.com/pypi/packages/56/15/064b197e855bfb7b343210e82490ae672f8bc7cdf3ddb02e92f64304ee8a/xxhash-3.6.0-cp313-cp313t-win_arm64.whl", hash = "sha256:ec44b73a4220623235f67a996c862049f375df3b1052d9899f40a6382c32d746" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "yarl"
Feat: chat channels — connect assistants to external messaging bots (#15850) ### What problem does this PR solve? #15844 Adds a **Chat channels** capability so a RAGFlow assistant (Dialog) can be exposed as a bot on external messaging platforms (Feishu/Lark, Discord, Telegram, Slack, WeCom, LINE, etc.). An admin configures a bot in the UI, connects it to an assistant, and inbound messages are answered from that assistant's knowledge base — replies are delivered back on the channel. **Feishu/Lark is implemented and tested end-to-end.** Discord, Telegram, LINE, and WeCom are scaffolded against the same interface; the remaining listed channels are tracked as follow-ups. ### Design **Backend** - New `chat_channel` table (`tenant_id`, `name`, `channel`, `config` JSON holding `{credential: {...}}`, `dialog_id`, `status`) + `ChatChannelService` and RESTful CRUD under `/api/v1/chat_channels`. - Channel framework under `api/channels/`: a `core` registry + per-channel packages that self-register a builder and implement a common `Channel` interface (`start`/`stop`/`send` + inbound normalization) over `IncomingMessage`/`OutgoingMessage`. - Embedded **reconcile loop** in `ragflow_server` (`api/channels/bootstrap.py`): loads enabled bots, and starts/stops/restarts them as rows change (no server restart needed). Inbound messages run the connected dialog via the non-streaming completion path, keeping per-end-user conversation history. - Missing optional channel SDKs degrade gracefully (channel skipped with a warning; others unaffected). Channel-level errors are logged, not crashed. - Feishu's WebSocket client runs in a dedicated thread with its own event loop to avoid cross-loop/contextvars conflicts with the channel runtime. **Frontend** - **Settings → Chat channels** panel: available-channels grid + configured-bots list with add/edit/delete and a **Connect assistant** popup that binds a bot to a dialog. - Brand icons via simple-icons / reused shared data-source assets, with colored fallbacks for brands not available. - Route, sidebar entry, i18n (en/zh), and a top-nav segment-boundary fix so the settings page no longer highlights the Chat tab. ### Type of change - [x] New Feature (non-breaking change which adds functionality) ### Notes - DB: new `chat_channel` table is auto-created; `chat_channel.dialog_id` is also covered by a `migrate_db` `alter_db_add_column` for existing installs. - Channel SDKs (`lark-oapi`, `discord.py`, `python-telegram-bot`, `line-bot-sdk`, `wechatpy`, `aiohttp`) added to dependencies. - Screenshots / per-channel credential docs to follow. <img width="1338" height="1290" alt="Image" src="https://github.com/user-attachments/assets/042cb2f9-0dad-4e6a-bcf7-43ced4bbd704" /> <img width="1344" height="738" alt="Image" src="https://github.com/user-attachments/assets/373cd08e-ec40-4c67-9c51-4d948b1ba617" /> <img width="672" height="887" alt="Image" src="https://github.com/user-attachments/assets/5a34953f-a9a3-4c1e-869e-5eff0dc64c84" /> ---------
2026-06-12 18:21:30 +08:00
version = "1.22.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "idna" },
{ name = "multidict" },
{ name = "propcache" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/57/63/0c6ebca57330cd313f6102b16dd57ffaf3ec4c83403dcb45dbd15c6f3ea1/yarl-1.22.0.tar.gz", hash = "sha256:bebf8557577d4401ba8bd9ff33906f1376c877aa78d1fe216ad01b4d6745af71" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/ea/f3/d67de7260456ee105dc1d162d43a019ecad6b91e2f51809d6cddaa56690e/yarl-1.22.0-cp313-cp313-macosx_10_13_universal2.whl", hash = "sha256:8dee9c25c74997f6a750cd317b8ca63545169c098faee42c84aa5e506c819b53" },
{ url = "https://mirrors.aliyun.com/pypi/packages/01/88/04d98af0b47e0ef42597b9b28863b9060bb515524da0a65d5f4db160b2d5/yarl-1.22.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:01e73b85a5434f89fc4fe27dcda2aff08ddf35e4d47bbbea3bdcd25321af538a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/18/91/3274b215fd8442a03975ce6bee5fe6aa57a8326b29b9d3d56234a1dca244/yarl-1.22.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:22965c2af250d20c873cdbee8ff958fb809940aeb2e74ba5f20aaf6b7ac8c70c" },
{ url = "https://mirrors.aliyun.com/pypi/packages/61/3a/caf4e25036db0f2da4ca22a353dfeb3c9d3c95d2761ebe9b14df8fc16eb0/yarl-1.22.0-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:b4f15793aa49793ec8d1c708ab7f9eded1aa72edc5174cae703651555ed1b601" },
{ url = "https://mirrors.aliyun.com/pypi/packages/6e/9e/51a77ac7516e8e7803b06e01f74e78649c24ee1021eca3d6a739cb6ea49c/yarl-1.22.0-cp313-cp313-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:e5542339dcf2747135c5c85f68680353d5cb9ffd741c0f2e8d832d054d41f35a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d4/f8/33b92454789dde8407f156c00303e9a891f1f51a0330b0fad7c909f87692/yarl-1.22.0-cp313-cp313-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:5c401e05ad47a75869c3ab3e35137f8468b846770587e70d71e11de797d113df" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d9/9a/c5db84ea024f76838220280f732970aa4ee154015d7f5c1bfb60a267af6f/yarl-1.22.0-cp313-cp313-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:243dda95d901c733f5b59214d28b0120893d91777cb8aa043e6ef059d3cddfe2" },
{ url = "https://mirrors.aliyun.com/pypi/packages/11/c9/cd8538dc2e7727095e0c1d867bad1e40c98f37763e6d995c1939f5fdc7b1/yarl-1.22.0-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:bec03d0d388060058f5d291a813f21c011041938a441c593374da6077fe21b1b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/a1/b9/ab437b261702ced75122ed78a876a6dec0a1b0f5e17a4ac7a9a2482d8abe/yarl-1.22.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:b0748275abb8c1e1e09301ee3cf90c8a99678a4e92e4373705f2a2570d581273" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b2/9d/8e1ae6d1d008a9567877b08f0ce4077a29974c04c062dabdb923ed98e6fe/yarl-1.22.0-cp313-cp313-musllinux_1_2_armv7l.whl", hash = "sha256:47fdb18187e2a4e18fda2c25c05d8251a9e4a521edaed757fef033e7d8498d9a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ca/5a/09b7be3905962f145b73beb468cdd53db8aa171cf18c80400a54c5b82846/yarl-1.22.0-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:c7044802eec4524fde550afc28edda0dd5784c4c45f0be151a2d3ba017daca7d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/aa/7f/59ec509abf90eda5048b0bc3e2d7b5099dffdb3e6b127019895ab9d5ef44/yarl-1.22.0-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:139718f35149ff544caba20fce6e8a2f71f1e39b92c700d8438a0b1d2a631a02" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e5/84/891158426bc8036bfdfd862fabd0e0fa25df4176ec793e447f4b85cf1be4/yarl-1.22.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:e1b51bebd221006d3d2f95fbe124b22b247136647ae5dcc8c7acafba66e5ee67" },
{ url = "https://mirrors.aliyun.com/pypi/packages/bb/49/03da1580665baa8bef5e8ed34c6df2c2aca0a2f28bf397ed238cc1bbc6f2/yarl-1.22.0-cp313-cp313-win32.whl", hash = "sha256:d3e32536234a95f513bd374e93d717cf6b2231a791758de6c509e3653f234c95" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9a/ee/450914ae11b419eadd067c6183ae08381cfdfcb9798b90b2b713bbebddda/yarl-1.22.0-cp313-cp313-win_amd64.whl", hash = "sha256:47743b82b76d89a1d20b83e60d5c20314cbd5ba2befc9cda8f28300c4a08ed4d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/98/4d/264a01eae03b6cf629ad69bae94e3b0e5344741e929073678e84bf7a3e3b/yarl-1.22.0-cp313-cp313-win_arm64.whl", hash = "sha256:5d0fcda9608875f7d052eff120c7a5da474a6796fe4d83e152e0e4d42f6d1a9b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/88/fc/6908f062a2f77b5f9f6d69cecb1747260831ff206adcbc5b510aff88df91/yarl-1.22.0-cp313-cp313t-macosx_10_13_universal2.whl", hash = "sha256:719ae08b6972befcba4310e49edb1161a88cdd331e3a694b84466bd938a6ab10" },
{ url = "https://mirrors.aliyun.com/pypi/packages/65/47/76594ae8eab26210b4867be6f49129861ad33da1f1ebdf7051e98492bf62/yarl-1.22.0-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:47d8a5c446df1c4db9d21b49619ffdba90e77c89ec6e283f453856c74b50b9e3" },
{ url = "https://mirrors.aliyun.com/pypi/packages/ab/ce/05e9828a49271ba6b5b038b15b3934e996980dd78abdfeb52a04cfb9467e/yarl-1.22.0-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:cfebc0ac8333520d2d0423cbbe43ae43c8838862ddb898f5ca68565e395516e9" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d1/c5/7dffad5e4f2265b29c9d7ec869c369e4223166e4f9206fc2243ee9eea727/yarl-1.22.0-cp313-cp313t-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:4398557cbf484207df000309235979c79c4356518fd5c99158c7d38203c4da4f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/50/b2/375b933c93a54bff7fc041e1a6ad2c0f6f733ffb0c6e642ce56ee3b39970/yarl-1.22.0-cp313-cp313t-manylinux2014_armv7l.manylinux_2_17_armv7l.manylinux_2_31_armv7l.whl", hash = "sha256:2ca6fd72a8cd803be290d42f2dec5cdcd5299eeb93c2d929bf060ad9efaf5de0" },
{ url = "https://mirrors.aliyun.com/pypi/packages/66/50/bfc2a29a1d78644c5a7220ce2f304f38248dc94124a326794e677634b6cf/yarl-1.22.0-cp313-cp313t-manylinux2014_ppc64le.manylinux_2_17_ppc64le.manylinux_2_28_ppc64le.whl", hash = "sha256:ca1f59c4e1ab6e72f0a23c13fca5430f889634166be85dbf1013683e49e3278e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/46/96/f3941a46af7d5d0f0498f86d71275696800ddcdd20426298e572b19b91ff/yarl-1.22.0-cp313-cp313t-manylinux2014_s390x.manylinux_2_17_s390x.manylinux_2_28_s390x.whl", hash = "sha256:6c5010a52015e7c70f86eb967db0f37f3c8bd503a695a49f8d45700144667708" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c1/42/8b27c83bb875cd89448e42cd627e0fb971fa1675c9ec546393d18826cb50/yarl-1.22.0-cp313-cp313t-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:9d7672ecf7557476642c88497c2f8d8542f8e36596e928e9bcba0e42e1e7d71f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/49/36/99ca3122201b382a3cf7cc937b95235b0ac944f7e9f2d5331d50821ed352/yarl-1.22.0-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:3b7c88eeef021579d600e50363e0b6ee4f7f6f728cd3486b9d0f3ee7b946398d" },
{ url = "https://mirrors.aliyun.com/pypi/packages/85/b4/47328bf996acd01a4c16ef9dcd2f59c969f495073616586f78cd5f2efb99/yarl-1.22.0-cp313-cp313t-musllinux_1_2_armv7l.whl", hash = "sha256:f4afb5c34f2c6fecdcc182dfcfc6af6cccf1aa923eed4d6a12e9d96904e1a0d8" },
{ url = "https://mirrors.aliyun.com/pypi/packages/c2/ad/b77d7b3f14a4283bffb8e92c6026496f6de49751c2f97d4352242bba3990/yarl-1.22.0-cp313-cp313t-musllinux_1_2_ppc64le.whl", hash = "sha256:59c189e3e99a59cf8d83cbb31d4db02d66cda5a1a4374e8a012b51255341abf5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/81/c8/06e1d69295792ba54d556f06686cbd6a7ce39c22307100e3fb4a2c0b0a1d/yarl-1.22.0-cp313-cp313t-musllinux_1_2_s390x.whl", hash = "sha256:5a3bf7f62a289fa90f1990422dc8dff5a458469ea71d1624585ec3a4c8d6960f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4b/b8/4c0e9e9f597074b208d18cef227d83aac36184bfbc6eab204ea55783dbc5/yarl-1.22.0-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:de6b9a04c606978fdfe72666fa216ffcf2d1a9f6a381058d4378f8d7b1e5de62" },
{ url = "https://mirrors.aliyun.com/pypi/packages/e0/e5/11f140a58bf4c6ad7aca69a892bff0ee638c31bea4206748fc0df4ebcb3a/yarl-1.22.0-cp313-cp313t-win32.whl", hash = "sha256:1834bb90991cc2999f10f97f5f01317f99b143284766d197e43cd5b45eb18d03" },
{ url = "https://mirrors.aliyun.com/pypi/packages/31/74/8b74bae38ed7fe6793d0c15a0c8207bbb819cf287788459e5ed230996cdd/yarl-1.22.0-cp313-cp313t-win_amd64.whl", hash = "sha256:ff86011bd159a9d2dfc89c34cfd8aff12875980e3bd6a39ff097887520e60249" },
{ url = "https://mirrors.aliyun.com/pypi/packages/69/66/991858aa4b5892d57aef7ee1ba6b4d01ec3b7eb3060795d34090a3ca3278/yarl-1.22.0-cp313-cp313t-win_arm64.whl", hash = "sha256:7861058d0582b847bc4e3a4a4c46828a410bca738673f35a29ba3ca5db0b473b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/73/ae/b48f95715333080afb75a4504487cbe142cae1268afc482d06692d605ae6/yarl-1.22.0-py3-none-any.whl", hash = "sha256:1380560bdba02b6b6c90de54133c81c9f2a453dee9912fe58c1dcced1edb7cff" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "yfinance"
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
version = "0.2.65"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
{ name = "beautifulsoup4" },
{ name = "curl-cffi" },
{ name = "frozendict" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "multitasking" },
{ name = "numpy" },
{ name = "pandas" },
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
{ name = "peewee" },
{ name = "platformdirs" },
{ name = "protobuf" },
{ name = "pytz" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
{ name = "requests" },
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
{ name = "websockets" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a3/c1/2ef5acda45a71297f4be22e205359e0f93b0171f2b6ebdd681362e725686/yfinance-0.2.65.tar.gz", hash = "sha256:3d465e58c49be9d61f9862829de3e00bef6b623809f32f4efb5197b62fc60485" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/c9/1e/631c80e0f97aef46eb73549b9b0f60d94057294e040740f4cad0cb1f48e4/yfinance-0.2.65-py2.py3-none-any.whl", hash = "sha256:7be13abb0d80a17230bf798e9c6a324fa2bef0846684a6d4f7fa2abd21938963" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "zhipuai"
version = "2.0.1"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
dependencies = [
{ name = "cachetools" },
{ name = "httpx" },
{ name = "pydantic" },
{ name = "pyjwt" },
]
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/a4/90/299e3456ee7ee1e118593552e03b86da2e9adaa0d454e467aeb4b22032a4/zhipuai-2.0.1.tar.gz", hash = "sha256:297bbdbe9393da2d1dc8066c39cf39bb2342f170d86f2b7b7a13ba368c53d701" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/8f/05/c3d4556886b5c6cf8c0b96eb80448ee8154c0dcc87086df018e817779ed4/zhipuai-2.0.1-py3-none-any.whl", hash = "sha256:738033d95696c3d5117dc4487e37d924e3ebbcdfa0072812b3f63a08ff72274a" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "zipp"
Feat: Redesign and refactor agent module (#9113) ### What problem does this PR solve? #9082 #6365 <u> **WARNING: it's not compatible with the older version of `Agent` module, which means that `Agent` from older versions can not work anymore.**</u> ### Type of change - [x] New Feature (non-breaking change which adds functionality)
2025-07-30 19:41:09 +08:00
version = "3.23.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/e3/02/0f2892c661036d50ede074e376733dca2ae7c6eb617489437771209d4180/zipp-3.23.0.tar.gz", hash = "sha256:a07157588a12518c9d4034df3fbbee09c814741a33ff63c05fa29d26a2404166" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/2e/54/647ade08bf0db230bfea292f893923872fd20be6ac6f53b2b936ba839d75/zipp-3.23.0-py3-none-any.whl", hash = "sha256:071652d6115ed432f5ce1d34c336c0adfd6a884660d1e9712a256d3d3bd4b14e" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "zlib-state"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "0.1.10"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/35/3e/dd482d5bf99d1dabcce0a20a479859cb7a6bd8a365b07b41ebf46b3c0f3d/zlib_state-0.1.10.tar.gz", hash = "sha256:c29b6b93cea1b80025fbc96fa91ceed8b5e7b54ef08f16d6e4c7f8fb56aad777" }
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
wheels = [
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
{ url = "https://mirrors.aliyun.com/pypi/packages/70/0c/2b0803cb9f30bddbc9eda87d251d958d21cfdde826bc1deb1e19ca0ff320/zlib_state-0.1.10-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:dfecba070cdeeab073573ac721459727d60e0b8ef7b38dac3c965459781b0eeb" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b1/d2/74ff59bb480801eae2731523f98be198eec135a9d37e27791b635f2c9124/zlib_state-0.1.10-cp313-cp313-win_amd64.whl", hash = "sha256:72e354f09c942055677ba59d76ca8c311a8129dfc98c3b44db33302843090204" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
[[package]]
name = "zstandard"
Import rag_tokenizer from Infinity (#11647) ### What problem does this PR solve? - Original rag/nlp/rag_tokenizer.py is put to Infinity and infinity-sdk via https://github.com/infiniflow/infinity/pull/3117 . Import rag_tokenizer from infinity and inherit from rag_tokenizer.RagTokenizer in new rag/nlp/rag_tokenizer.py. - Bump infinity to 0.6.8 ### Type of change - [x] Refactoring
2025-12-02 14:59:37 +08:00
version = "0.25.0"
feat(go-agent): Ported retrieval node, added Keenable web search tool (#16396) Ported retrieval node, added Keenable web search tool - [x] New Feature (non-breaking change which adds functionality)
2026-06-26 22:55:49 +08:00
source = { registry = "https://mirrors.aliyun.com/pypi/simple" }
sdist = { url = "https://mirrors.aliyun.com/pypi/packages/fd/aa/3e0508d5a5dd96529cdc5a97011299056e14c6505b678fd58938792794b1/zstandard-0.25.0.tar.gz", hash = "sha256:7713e1179d162cf5c7906da876ec2ccb9c3a9dcbdffef0cc7f70c3667a205f0b" }
wheels = [
{ url = "https://mirrors.aliyun.com/pypi/packages/35/0b/8df9c4ad06af91d39e94fa96cc010a24ac4ef1378d3efab9223cc8593d40/zstandard-0.25.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:ec996f12524f88e151c339688c3897194821d7f03081ab35d31d1e12ec975e94" },
{ url = "https://mirrors.aliyun.com/pypi/packages/3f/06/9ae96a3e5dcfd119377ba33d4c42a7d89da1efabd5cb3e366b156c45ff4d/zstandard-0.25.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:a1a4ae2dec3993a32247995bdfe367fc3266da832d82f8438c8570f989753de1" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d9/14/933d27204c2bd404229c69f445862454dcc101cd69ef8c6068f15aaec12c/zstandard-0.25.0-cp313-cp313-manylinux2010_i686.manylinux2014_i686.manylinux_2_12_i686.manylinux_2_17_i686.whl", hash = "sha256:e96594a5537722fdfb79951672a2a63aec5ebfb823e7560586f7484819f2a08f" },
{ url = "https://mirrors.aliyun.com/pypi/packages/6d/db/ddb11011826ed7db9d0e485d13df79b58586bfdec56e5c84a928a9a78c1c/zstandard-0.25.0-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:bfc4e20784722098822e3eee42b8e576b379ed72cca4a7cb856ae733e62192ea" },
{ url = "https://mirrors.aliyun.com/pypi/packages/db/00/87466ea3f99599d02a5238498b87bf84a6348290c19571051839ca943777/zstandard-0.25.0-cp313-cp313-manylinux2014_ppc64le.manylinux_2_17_ppc64le.whl", hash = "sha256:457ed498fc58cdc12fc48f7950e02740d4f7ae9493dd4ab2168a47c93c31298e" },
{ url = "https://mirrors.aliyun.com/pypi/packages/2b/95/fc5531d9c618a679a20ff6c29e2b3ef1d1f4ad66c5e161ae6ff847d102a9/zstandard-0.25.0-cp313-cp313-manylinux2014_s390x.manylinux_2_17_s390x.whl", hash = "sha256:fd7a5004eb1980d3cefe26b2685bcb0b17989901a70a1040d1ac86f1d898c551" },
{ url = "https://mirrors.aliyun.com/pypi/packages/63/4b/e3678b4e776db00f9f7b2fe58e547e8928ef32727d7a1ff01dea010f3f13/zstandard-0.25.0-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:8e735494da3db08694d26480f1493ad2cf86e99bdd53e8e9771b2752a5c0246a" },
{ url = "https://mirrors.aliyun.com/pypi/packages/4e/d5/ba05ed95c6b8ec30bd468dfeab20589f2cf709b5c940483e31d991f2ca58/zstandard-0.25.0-cp313-cp313-musllinux_1_1_aarch64.whl", hash = "sha256:3a39c94ad7866160a4a46d772e43311a743c316942037671beb264e395bdd611" },
{ url = "https://mirrors.aliyun.com/pypi/packages/50/d5/870aa06b3a76c73eced65c044b92286a3c4e00554005ff51962deef28e28/zstandard-0.25.0-cp313-cp313-musllinux_1_1_x86_64.whl", hash = "sha256:172de1f06947577d3a3005416977cce6168f2261284c02080e7ad0185faeced3" },
{ url = "https://mirrors.aliyun.com/pypi/packages/5d/35/398dc2ffc89d304d59bc12f0fdd931b4ce455bddf7038a0a67733a25f550/zstandard-0.25.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:3c83b0188c852a47cd13ef3bf9209fb0a77fa5374958b8c53aaa699398c6bd7b" },
{ url = "https://mirrors.aliyun.com/pypi/packages/9a/5c/36ba1e5507d56d2213202ec2b05e8541734af5f2ce378c5d1ceaf4d88dc4/zstandard-0.25.0-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:1673b7199bbe763365b81a4f3252b8e80f44c9e323fc42940dc8843bfeaf9851" },
{ url = "https://mirrors.aliyun.com/pypi/packages/70/e8/2ec6b6fb7358b2ec0113ae202647ca7c0e9d15b61c005ae5225ad0995df5/zstandard-0.25.0-cp313-cp313-musllinux_1_2_ppc64le.whl", hash = "sha256:0be7622c37c183406f3dbf0cba104118eb16a4ea7359eeb5752f0794882fc250" },
{ url = "https://mirrors.aliyun.com/pypi/packages/7b/01/b5f4d4dbc59ef193e870495c6f1275f5b2928e01ff5a81fecb22a06e22fb/zstandard-0.25.0-cp313-cp313-musllinux_1_2_s390x.whl", hash = "sha256:5f5e4c2a23ca271c218ac025bd7d635597048b366d6f31f420aaeb715239fc98" },
{ url = "https://mirrors.aliyun.com/pypi/packages/b2/e5/fbd822d5c6f427cf158316d012c5a12f233473c2f9c5fe5ab1ae5d21f3d8/zstandard-0.25.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:4f187a0bb61b35119d1926aee039524d1f93aaf38a9916b8c4b78ac8514a0aaf" },
{ url = "https://mirrors.aliyun.com/pypi/packages/8e/e0/69a553d2047f9a2c7347caa225bb3a63b6d7704ad74610cb7823baa08ed7/zstandard-0.25.0-cp313-cp313-win32.whl", hash = "sha256:7030defa83eef3e51ff26f0b7bfb229f0204b66fe18e04359ce3474ac33cbc09" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d9/82/b9c06c870f3bd8767c201f1edbdf9e8dc34be5b0fbc5682c4f80fe948475/zstandard-0.25.0-cp313-cp313-win_amd64.whl", hash = "sha256:1f830a0dac88719af0ae43b8b2d6aef487d437036468ef3c2ea59c51f9d55fd5" },
{ url = "https://mirrors.aliyun.com/pypi/packages/d4/57/60c3c01243bb81d381c9916e2a6d9e149ab8627c0c7d7abb2d73384b3c0c/zstandard-0.25.0-cp313-cp313-win_arm64.whl", hash = "sha256:85304a43f4d513f5464ceb938aa02c1e78c2943b29f44a750b48b25ac999a049" },
Replace poetry with uv (#4471) ### What problem does this PR solve? Replace poetry with uv ### Type of change - [x] Refactoring
2025-01-14 11:49:43 +08:00
]
Reference in New Issue Copy Permalink