eb6c2b8e2a
Merge branch 'feat/model-finetune-opt' into feat/cli-skill-sync
2026-08-07 14:26:38 +08:00
3c64461cca
feat: video finetune/deploy/invoke full pipeline + training cost calculation
...
- Add finetune video create subcommand (wan2.7/2.5/2.2 i2v + kf2v)
- Align video hyperparams with official docs (n_epochs=50, per-model batch_size/max_pixels)
- Add --last-frame flag to video generate for kf2v (image2video endpoint)
- Fix wan2.1-2.6 i2v input format (img_url instead of media[])
- Add training_cost field to finetune get/watch (catalog ft price, API-key domain only)
- Add --aigc-* flags to deploy create (optional, for video LoRA prompt config)
2026-08-07 07:03:01 +08:00
24092b423c
fix(usage): handle unavailable token plan quotas
2026-08-06 22:47:26 +08:00
f30fff9065
feat(finetune): clarify model flags; add price estimate & actual cost
...
- Rename for clarity: finetune --model → --base-model (create/price/
capability/list); deploy create --model → --model-name, --name →
--display-name
- Add `finetune price` (console domain) for pre-training cost estimate
(sft/dpo/cpt)
- Add actual training cost (fee.ts) enriched into finetune get/watch
from catalog price × reported usage
2026-08-06 15:08:02 +08:00
752a79e442
fix(usage): handle missing token plan reset times
2026-08-06 09:12:41 +08:00
80bdcb83f6
feat(usage): add token plan usage view
2026-08-05 23:28:39 +08:00
d9e8601a50
feat(knowledge): 支持上传目录路径并递归扫描文件
...
- 支持上传参数中传入目录路径，递归扫描子目录下文件
- 自动忽略 node_modules、.git 等常见工具目录
- 不支持的文件格式不会报错，跳过并列表提示
- 上传时校验扩展名和大小限制，支持批量文件上传
- 输出中增加跳过的文件列表，verbose 模式下显示详细文件名
- 测试覆盖目录上传、文件跳过和空目录等场景
- 更新相关文档，说明新支持的目录上传功能及注意事项
2026-08-05 22:29:39 +08:00
e3bb5a7fa0
test(commands): 添加知识库统计接口的E2E测试
...
- 在topic-routes中新增knowledge stats路由映射
- 在j2-content-ops测试中添加知识库统计命令调用
- 验证接口返回的存储限制和请求速率监控数据结构有效
- 记录存储限制和请求窗口数量作为测试备注
- 引入parseStdoutJson辅助函数解析JSON输出
2026-08-05 20:13:42 +08:00
ef463e8d5d
test(e2e): 优化检索结果标记召回判断并完善服务调优用例
...
- 新增节点召回标记判断函数 nodesRecallMarker，避免误判 marker 出现位置
- 将所有相关轮询断言替换为基于 nodesRecallMarker 的更严格判断
- 调整删除测试中对误伤判断的断言逻辑，确保准确检测召回标记
- 扩展服务调优用例，增加描述和温度参数调优测试，验证配置持久化
- 添加通过配置文件更新 kb_search_configs 并校验嵌套配置修改生效
- 部署后验证发布版本配置正确包含所有调优项
- 更新流程注释与断言提示，提升测试用例可读性和覆盖度
2026-08-05 19:56:52 +08:00
8ee2c378f5
test(knowledge): 添加多模态与表格型仓库 E2E 测试套件支持
...
- 新增多模态问答及检索服务的 live E2E 测试用例，支持基于图像参数的功能验证
- 补充表格库的 chunk add/list/delete 闭环测试，验证了 field channel 的必填项和读写一致性
- 增加图片库 chunk list 的 metadata 验证，确保 image_url 数组和可见性标志存在
- 实现带覆盖重导功能的 doc import-oss 测试，确认覆盖后 fileId 变更及旧文件失效
- 编写自有 OSS Bucket 的幂等复用集合创建和获取测试，确保服务端的 tag-based 访问控制支持
- 在 gating 中添加对各类长驻知识库及服务环境变量的就绪检测函数
2026-08-05 18:14:23 +08:00
4990b27436
feat: update image default model
2026-08-05 16:58:29 +08:00
43abf0aca5
feat(knowledge): 新增知识库管理及用户旅程端到端测试支持
...
- 增加test:journey脚本，覆盖知识库跨命令全链路用户旅程测试
- 在文档中新增Journey E2E章节，详细说明用户旅程测试定位及断言机制
- 完善commands模块，新增知识库相关命令包括知识库列表、信息、创建、更新、删除
- 新增知识库文档相关命令，如文档列表、状态、上传、删除、打标签及OSS导入
- 添加知识服务管理命令，支持列表、创建、更新、部署、删除及复制
- 支持知识块增删查改命令，完善知识点的灵活操作能力
- 实现数据中心分类管理命令，支持分类增删查操作
- 优化knowledge chat命令，增加workspace-id统一解析及agent-version版本控制
- 重构与知识库相关命令的导出与注册，完善CLI整体能力覆盖
- 新增命令详尽的帮助文档，包含参数说明、使用示例及错误边界
- 实现批量删除知识块的自动分批处理逻辑，易于操作大规模数据
- 添加必要的输入校验与安全提示，确保操作安全且符合规范
2026-08-05 12:02:32 +08:00
081d09863b
Merge branch 'main' into feat/multi-channel-install
2026-08-04 20:22:26 +08:00
3766b6d7ca
Merge branch 'main' into feat/bailian-wiki-doc-sync
2026-08-03 19:33:16 +08:00
026e250cd3
Merge branch 'main' into feat/multi-channel-install
2026-08-03 17:16:56 +08:00
1c38c381e5
feat: switch default text model to qwen3.8-max
...
Align text chat, pipeline, config UI, login validation, and Token Plan
text presets, and update README, skill reference, and related tests.
2026-08-03 15:34:18 +08:00
658763af2c
fix: ci test
2026-08-03 15:29:55 +08:00
9fb388b75d
Merge branch 'main' of github.com:modelstudioai/cli into feat/multi-channel-install
2026-07-31 17:47:10 +08:00
45d468838f
feat: win bl update exe file test
2026-07-31 17:44:59 +08:00
30fe8182f4
Merge branch 'main' into feat/bailian-wiki-doc-sync
...
# Conflicts:
# packages/cli/src/commands.ts
# packages/commands/tests/e2e/topic-routes.ts
# pnpm-lock.yaml
# pnpm-workspace.yaml
# skills/bailian-cli/reference/index.md
2026-07-29 17:34:28 +08:00
65c0fe9604
feat: add skill commend & skill install
2026-07-29 17:04:17 +08:00
2c53b0692b
refactor(inventory): 优化技能与代理配置代码格式和检测逻辑
...
- 统一代码格式，增加多处代码块的换行和缩进保持一致
- 调整技能安装目标列表的格式，提升可读性
- 修复解压缩逻辑中异常抛出格式，增强异常信息规范
- 优化归一化文件名过滤条件表达式格式
- 修改配置文件检测逻辑，兼容环境变量和旧版配置方案
- 增强对 Bailian 相关模型提供者的检测逻辑支持
- 规范代理详情字段生成方法的代码风格
- 调整 MCP 写回相关函数的格式，提升可维护性
- 改进技能和代理详情函数参数格式，统一参数拆分显示
- 修复单元测试中路径和 JSON 写入格式，增加不同配置场景测试覆盖
- 确保软链接技能目录被正确识别为安装来源
- 增加多代理配置文件和技能安装的检测测试用例，提升测试精准度
2026-07-28 20:51:07 +08:00
adc89f635d
Merge branch 'main' of github.com:modelstudioai/cli into feat/config-ui-enhancements
...
# Conflicts:
# packages/commands/tests/config-ui.test.ts
2026-07-28 20:43:42 +08:00
bbf45a5961
Merge branch 'main' of github.com:modelstudioai/cli into feat/config-agent-fix
...
# Conflicts:
# CHANGELOG.md
# CHANGELOG.zh.md
# packages/cli/package.json
# packages/commands/package.json
# packages/core/package.json
# packages/kscli/package.json
# packages/runtime/package.json
# skills/bailian-cli/SKILL.md
2026-07-28 20:33:16 +08:00
96744e3328
feat(config-agent): add --key and --region, default codex wire_api to responses
...
- --key: decode the web console's obfuscated API key (o1_ prefix) into
the real key; mutually exclusive with --api-key, exactly one required
- --region: convert a Model Studio region into the Token Plan base URL
(token-plan.<region>.maas.aliyuncs.com/compatible-mode/v1); mutually
exclusive with --base-url, exactly one required
- codex: default wire_api to "responses" (current Codex rejects "chat");
--wire-api chat kept for legacy Codex <= 0.80.0 with a warning
- regenerate skills reference for the new flags
2026-07-28 19:37:13 +08:00
25ac5c9c84
fix: revert change about defaultTextModel
2026-07-28 18:05:18 +08:00
5d9e22de8f
Merge remote-tracking branch 'refs/remotes/origin/main' into fix/web-search-and-thinking
2026-07-28 17:20:46 +08:00
20e3555b84
fix(text,auth): drop enable_thinking retry and omit the field by default
...
Pass through model constraint errors instead of auto-retrying, switch token-plan default text model to qwen3.7-plus, and align related e2e expectations.
2026-07-28 17:19:05 +08:00
17c52fb86f
Merge branch 'main' into fix/default-model
2026-07-28 15:46:46 +08:00
7ad14a79b9
Merge branch 'main' of github.com:modelstudioai/cli into fix/web-search-and-thinking
2026-07-28 09:25:34 +08:00
8211268bd8
fix(text,mcp): keep thinking_budget on enable_thinking retry and hint MCP activation on 404
2026-07-28 09:25:02 +08:00
2dce9fe093
feat(agent): session and destroy failed error
2026-07-27 21:52:52 +08:00
9819eb6ddc
feat(agent): 非bailian provider也走鉴权逻辑
2026-07-27 21:33:36 +08:00
a03ee0c72c
fix(agent): timeout error
2026-07-27 20:23:21 +08:00
05860b3bdd
fix(agent): session output json with session_id
2026-07-27 20:10:47 +08:00
63ee5aaec3
fix(agent): plan command dry-run
2026-07-27 20:03:09 +08:00
93c9149e45
feat(agent): 鉴权分离线命令和在线命令，仅对bailian provider鉴权
2026-07-27 19:51:56 +08:00
0221e35803
fix(config-agent): 优化 Codex 配置写入与兼容性处理
...
- 调整 Codex 代理默认 wire_api 为 "responses"，兼容新版 Codex
- 增加对 legacy Codex <= 0.80.0 使用 wire_api "chat" 的警告提示
- 修正 agent flags 描述，更准确说明 wire_api 默认与兼容范围
- 优化代码格式，统一 import 语句风格
- 增加测试用例覆盖不同 wire_api 配置及环境变量警告
- 修复写入过程中文件备份及合并逻辑，保留用户已有配置
- 修复多个 provider 写入时键名与内容匹配，避免重复添加
- 改善测试代码格式，提高可读性与一致性
2026-07-27 17:10:26 +08:00
36ebd63716
fix(text): omit enable_thinking by default and retry when API requires false
...
Non-streaming chat no longer forces enable_thinking=false, which breaks
thinking-only models. Retry once with false only when the server demands it.
2026-07-27 16:51:11 +08:00
dac254af86
feat: add MCP WebSearch page URL and enhance error handling in web search command
2026-07-27 16:51:09 +08:00
8a0fb870f1
feat(agent): update openagentpack sdk version
2026-07-27 16:34:58 +08:00
ff469ce717
feat(install-docs): enhance installation documentation and validation processes
2026-07-27 11:24:22 +08:00
5f0966ec8d
fix(agent): ci issues
2026-07-27 10:51:09 +08:00
c4f5bb09c6
fix(image): resolve size and prompt_extend by model profile
...
Stop inferring size/prompt_extend from sync vs async; use per-family sizeProfile. wanx*-imageedit uses function+base_image_url; bare qwen-image uses the fixed resolution table.
2026-07-27 10:17:09 +08:00
9a13700390
fix(image): route text-to-image and image-edit by model family
...
Fix wanx/wan2.x-t2i, wan2.5-i2i, z-image, and qwen-image-plus hitting the wrong endpoint, and add routing unit tests plus dry-run coverage.
2026-07-27 10:16:25 +08:00
32c497db63
feat(agent): add skill-list command and fix pr issues
2026-07-26 18:52:27 +08:00
247bb82154
test(agent): cover config-write, profile, logout and error-mapping auth-chain scenarios
2026-07-24 18:44:18 +08:00
1bf4fec9e6
feat(agent): support --dry-run for all local and remote mutations
2026-07-24 16:02:15 +08:00
1d589c5178
Merge remote-tracking branch 'origin/main' into feat/cma
2026-07-24 14:52:21 +08:00
9cad1994e7
feat(agent): validate by client apiKey auth type
2026-07-24 14:50:30 +08:00
故璃