* Updating skilltoolset page per issue #1350 - 4
- Added a note so the developer checks for the files in their working directory.
- Removed redundant and not useful information for the developer on the agent's original
pr.
- Small formatting to headers.
Add a featured "Looker Block" subsection to "Build a dashboard" pointing to the
ready-to-use BigQuery Agent Analytics Looker Block on the Looker Marketplace,
and move the existing example-notebook guidance under a "Custom dashboard from a
notebook" subsection.
Variables app_name and user_id were defined lowercase but referenced as
APP_NAME and USER_ID (uppercase), causing NameError at runtime.
Co-authored-by: Kristopher Overholt <koverholt@google.com>
* VertexAiSearchTool dynamic config according to issue #1227 - 7
Added dynamic configuration per the bot's suggestion, but I improved snippet and description.
* Update agent-search.md
* Align snippet param name with SDK signature (readonly_context)
---------
Co-authored-by: Kristopher Overholt <koverholt@google.com>
* Update index.md
-Added the authentication note
-Did clean up on headers and the notes and warning format
* Document GOOGLE_API_KEY auth for Vertex AI eval criteria; clean up evaluate/index.md
---------
Co-authored-by: Kristopher Overholt <koverholt@google.com>
* Add anthropic thinking blocks and adaptive thinking according to issue #1521 - 20.md
-Added the thinking blocks feature to the documentation
-Additionally I added the adaptive thinking that applies for newer models.
* Reorganize Claude models adaptive thinking docs; add pointers to using Claude models with Python
* Remove in-line version references
---------
Co-authored-by: Kristopher Overholt <koverholt@google.com>
* docs(evaluate): document rubric_based_multi_turn_trajectory_quality_v1 and clarify Rubric.type / criterion.rubrics
Closes three gaps in docs/evaluate/criteria.md:
1. Adds a section for `rubric_based_multi_turn_trajectory_quality_v1`. The
metric is registered in metric_evaluator_registry.py and implemented in
rubric_based_multi_turn_trajectory_evaluator.py upstream, but the
explanatory section that its two sibling rubric-based metrics have was
missing. Also adds the corresponding row to the criteria table.
2. Adds a "Notes On Rubrics" subsection under each of the three
rubric-based metric sections, stating the required Rubric.type value
(FINAL_RESPONSE_QUALITY / TOOL_USE_QUALITY / TRAJECTORY_QUALITY) and the
criterion-vs-EvalCase rubric relationship (criterion.rubrics required
and non-empty; EvalCase.rubrics is additive, filtered by type).
3. Adds the explicit `type` field to existing JSON examples in the two
existing rubric-based sections so users have a complete template.
Refs #1852.
* docs(criteria): correct Rubric.type filter scope in the notes
Only EvalCase.rubrics is filtered by type in create_effective_rubrics_list;
criterion-level rubrics on EvalConfig are added unconditionally. Update the
Notes On Rubrics bullet in each of the three rubric-based metric sections
to reflect this, using koverholt's review suggestion.
* docs(criteria): drop inert type field from criterion-level rubric examples
type is only consulted for entries in EvalCase.rubrics, per
create_effective_rubrics_list. Keeping it on criterion-level rubrics
implied they were type-filtered, which they are not. Removed from all
six criterion-level rubric entries across the three rubric-based metric
sections.
* docs(criteria): add EvalCase.rubrics snippets showing where type actually applies
The type field only affects filtering on EvalCase.rubrics per
create_effective_rubrics_list. Show that behavior concretely in each of
the three rubric-based metric sections by adding a small EvalCase.rubrics
snippet with the type field set to the criterion's expected value, plus a
one-line reminder that the effective rubric set passed to the judge is the
union of criterion-level rubrics and any type-matching entries from
EvalCase.rubrics.
---------
Co-authored-by: Kristopher Overholt <koverholt@google.com>
* Update llm-agents.md
- Added adk-bot's suggestion
- Improved the readability of the snippet
- Did minor style corrections
* Update llm-agents.md
noticed a broken link and fixed it
* Fix set_default_model code block formatting and trailing whitespace
* Clean up llm-agents formatting, fix related links, and add version to set_default_model section
* Simplify code snippet
---------
Co-authored-by: Kristopher Overholt <koverholt@google.com>
* Add Unstructured Transform MCP integration
* Add pacing helper, env setup, and Gemini key prerequisite from review
* Re-trigger CLA check after signing
* Use the blue Unstructured logo for the integration icon
* Note that the agent can retry a weak parse with the hi_res strategy
* formatting changes
* Adopt nested tab convention, make wait helper async, replace hyphen asides
---------
Co-authored-by: Kristopher Overholt <koverholt@google.com>
* docs: Document new add_memory method for direct memory management
* 1350-6Memoryupdate.md
* Update on memory.md according to issue #1350 - 6
Simplified dense sections, corrected markdown formatting errors, and fixed typos to align with documentation standards. Corrected some sections from the original page and the agent's PR that were a little clunky. Erased sentences inside paragraph's ,Joe's request, and included commas instead. All this while including the necessary update.
* Update memory.md
-Worked on feedback, fixed errors, changed wording slightly, and reviewed format.
-Some changes were lost when I tried to resolve conflicts between branches in my last commit. I added them back here.
* Fix add_memory example and memory operations list in memory.md
* Format memory.md: line wrapping, snippet cleanup, and Java example
---------
Co-authored-by: adk-bot <adk-bot@google.com>
Co-authored-by: Kristopher Overholt <koverholt@google.com>
* Update cloud-trace.md to include captured attributes
* Cloud trace small updates
Small change to the Cloud Trace section
* Update cloud-trace.md
* Update cloud-trace.md
* Update cloud trace page according to issue #1256 - 5
Managed the conflicts with the agent's PR. Updated the Google Cloud Console link. It would be more helpful to deep-link directly to the Trace Explorer rather than the root dashboard, so this has been adjusted. Added a brief introductory paragraph to the ## Resources section to provide better context for the provided links
* Update cloud-trace.md
Worked on feedback: fixed capitalization, ensured consistent formatting, removed extra spaces, and made the update more technically accurate.
* Update cloud-trace.md
Worked on missing feedback comments.
* Remove user.id from captured attributes and clarify heading in cloud-trace.md
* Format cloud-trace.md: line wrapping, tab spacing, and copy cleanup
---------
Co-authored-by: adk-bot <adk-bot@google.com>
Co-authored-by: Kristopher Overholt <koverholt@google.com>
* Updated Context injection according to issue #1438 - 17
This PR adds documentation for context injection.
* Update function-tools.md
Worked on feedback.
- Updated Imports: Fixed the import path for ToolContext to correctly pull from google.adk.tools instead of google.adk.types.
- Snippet Formatting: Removed the unnecessary parentheses around "secondary_agent" in the commented-out transfer_to_agent action example.
- Removed Dead Code: Removed the redundant pass statement at the end of the my_tool function.
* Fix context injection docs and headings in function-tools.md
* Format function-tools.md: line wrapping, bullet markers, and copy cleanup
---------
Co-authored-by: Kristopher Overholt <koverholt@google.com>
* docs(agents): scaffold managed agents page and nav entry
* docs(agents): add managed agents get-started and decision guidance
* docs(agents): complete managed agents page (how it works, limitations, resources)
* docs(agents): simplify managed agents "how it works" and refine copy
Replace the internals-heavy "How it works" paragraph with a high-level
description that links to the Interactions API, drop the scenario table
in favor of the two-bullet comparison, add a note on how ManagedAgent
maps to the Managed Agents API (data plane vs. control plane), and
remove em-dashes throughout.
* docs(agents): point Interactions API link to the Gemini API overview
* Clean up managed agents page naming, structure, bullets, and links, and add it to the agents overview
---------
Co-authored-by: Kristopher Overholt <koverholt@google.com>
* Add Sprites integration
Adds the Sprites integration page and catalog icon. Sprites (https://sprites.dev)
are persistent, stateful Linux sandboxes from Fly.io with checkpoint/restore,
via the sprites-adk plugin (pip install sprites-adk).
* Refine Sprites page: benefit-forward use cases, precise wording
* Address review: functional heading + runner registration in example
---------
Co-authored-by: Kristopher Overholt <koverholt@google.com>
* Enhance BigQuery Agent Analytics with new options
Added options for OpenTelemetry correlation, custom metadata capture, and payload column projection in BigQuery Agent Analytics configuration.
* Fix formatting of Interactive Demo Notebook link
fix format
* docs(bigquery-agent-analytics): replace version callouts with language support tags
Address review feedback (koverholt) on PR #1942: the page used ad-hoc,
non-conventional version callouts (per-row table tags, footnotes, version
suffixes in headers, "post-vX.Y.Z" notes) that are unmaintainable and, in
places, stale.
- Drop the "(v2.4.0+)" / "(1.27.0+)" suffixes from config-table rows and
section headers; convey minimum versions with the repo-standard
language-support-tag blocks on the relevant sections instead
(Trace correlation → Python v2.4.0, Automatically Created Views →
Python v1.27.0, ADK 2.0 workflow events → Python v2.3.0).
- Remove the "‡ Added in v2.4.0" and "† ... not in v2.2.0" view-table
footnotes; the ADK 2.0 views/columns now point to the ADK 2.0 section.
- Remove the stale top "Version Requirement" note (v2.2.0 is no longer the
latest release; ADK 2.0 workflow support shipped in v2.3.0) and the stale
ADK 2.0 section note; de-version the intro prose.
- Remove the "Structured tool declarations (post-v2.4.0)" note: that shape is
not in any tagged release (v2.4.0 emits tools as a list of names); it will
be re-added under a language-support-tag once it ships.
- Trim the Trace-correlation section so it no longer restates the option table,
keeping only the cross-option constraints and the combined example.
* #1029 - 8 vector_store_similarity_search and SpannerVectorStoreSettings
Added vector_store_similarity_search to SpannerToolset. SpannerVectorStoreSettings is introduced to support this feature.
* Update spanner.md
* Update spanner.md
Applied feedback. Additionally, reviewed the phrasing in the intro for clarity and concision and did some minor formatting edits.
* Fix wrapping, missing import, trailing whitespace, formatting fixes
* Restructure Spanner page: add intro, section headings, reorder vector search
* Add experimental tags/notes to corresponding integration pages
---------
Co-authored-by: Kristopher Overholt <koverholt@google.com>
* docs(integrations): rename Mimir Memory -> Perseus Vault Memory (supersedes #1877)
* docs(integrations): address review on perseus-vault memory page
- Restore "Use cases" section
- Add intro phrases before Installation and Use-with-agent code blocks
- Point the binary prereq at the Installation section
- Drop the non-existent `cargo install perseus-vault` step
- Remove the "(formerly Mimir/Mneme)" lineage note
- Add trailing newline
* docs(integrations): restore Perseus live context (optional) section
Carry over the optional [perseus] extra / perseus_context_agent section
from the original page, updated to the Perseus Vault names.
---------
Co-authored-by: Kristopher Overholt <koverholt@google.com>
* docs: update ADK Go imports to v2 for core concepts and tutorials
* docs: update ADK Go imports and APIs to v2 for runtime and safety
* docs: update ADK Go imports to v2 for features (workflows, tools, skills, sessions, plugins)
* docs: update ADK Go imports to v2 for operations and integrations
---------
Co-authored-by: Kristopher Overholt <koverholt@google.com>
The memory docs and the express-mode weather agent notebook referenced
`adk.tools.preload_memory_tool.PreloadMemoryTool()`, which raises
`AttributeError: module 'google.adk.tools' has no attribute
'preload_memory_tool'` because the submodule is not exposed via the
package's lazy `__getattr__`.
Use the documented public API `from google.adk.tools import preload_memory`
and pass the `preload_memory` instance directly, consistent with how the
sibling `load_memory` tool is shown in the same page.
Fixes#1771
Co-authored-by: Kristopher Overholt <koverholt@google.com>
* Add Security and deployment section to Visual Builder page (#1492)
Documents two behaviors introduced in ADK Python v1.27.4: builder endpoints
are only registered when the web UI is enabled (adk web), and Visual Builder
uploads are restricted to .yaml/.yml files. Verified against
src/google/adk/cli/fast_api.py at current main.
* Issue #1492 -1 Visual Builder Security and deployment
Expanded Project Structure Documentation: Added a detailed, itemized breakdown explaining the specific role and functionality of each generated file and directory within the DiceAgent/ project structure.
Added Security & Deployment Guidelines: Introduced a new section outlining local API endpoint behaviors, headless/production constraints, and strict file upload validation rules (such as extension whitelisting and path traversal blocks) to align the documentation with the latest security updates in the codebase.
* Document builder endpoint security and deployment behavior
* Formatting edits
---------
Co-authored-by: Kristopher Overholt <koverholt@google.com>
* docs(integrations): add CarsXE MCP integration
Add CarsXE to the integrations catalog as an MCP server integration.
The hosted CarsXE MCP server (https://mcp.carsxe.com/mcp, streamable
HTTP) exposes vehicle data — VIN specs, plate decoding, market value,
history, recalls, lien/theft, OBD codes, images — as MCP tools wired
into an ADK agent via McpToolset. Includes Python and TypeScript
examples, the full tool list, and an icon asset.
* Update docs/integrations/carsxe.md
Co-authored-by: Kristopher Overholt <koverholt@google.com>
* Update docs/integrations/carsxe.md
Co-authored-by: Kristopher Overholt <koverholt@google.com>
---------
Co-authored-by: Kristopher Overholt <koverholt@google.com>
The "Runner start callbacks" section described `before_run_callback` as
returning a `types.Content` to "replace the user's original message". That is
actually the behavior of `on_user_message_callback`; the text (including the
"Purpose" line) was duplicated from that section.
In the implementation, returning a `Content` from `before_run_callback` halts
the run: the Runner yields it as an early-exit event and skips all subsequent
processing instead of invoking the agent.
See `Runner._exec_with_plugin` in `src/google/adk/runners.py`:
early_exit_result = await plugin_manager.run_before_run_callback(...)
if isinstance(early_exit_result, types.Content):
early_exit_event = Event(..., content=early_exit_result)
...
yield early_exit_event # early exit, agent never runs
else:
# normal execution
This also matches the BasePlugin docstring: "Returning a value to halt
execution of the runner and ends the runner with that event."
Fix the "When It Runs", "Purpose", and "Flow Control" bullets accordingly.
Co-authored-by: Kristopher Overholt <koverholt@google.com>
* I updated the section to include token id
* Update authentication.md
* Update authentication.md
* Update authentication.md
* Update authentication.md
* Update authentication.md
* Fix heading casing and anchor link text in ID token section
* Drop letter prefixes from toolset config headings
* Expand intro to "Configure tools with authentication"
---------
Co-authored-by: Kristopher Overholt <koverholt@google.com>
Now that the Perseus Context integration page is live, point to it
instead of the raw GitHub repo, mirroring how perseus.md already
links back to /integrations/mimir/.
Co-authored-by: Kristopher Overholt <koverholt@google.com>
Zyan