chore(release/candidate): release 1.27.0

[Jacksunwei]

🤖 I have created a release beep boop

1.27.0 (2026-03-11)

Features

• Add adk optimize command (b18d7a1)
• Add code execution in the sample agent (a86aa8b)
• add Dataplex Catalog search tool to BigQuery ADK (82c2eef)
• Add gcs filesystem support for Skills. It supports skills in text and pdf format, also has some sample agents (6edcb97)
• Add GEPA root agent prompt optimizer (4e3e2cb)
• Add interface between optimization infra and LocalEvalService (7b7ddda)
• Add list_skills_in_dir to skills utils (327b3af)
• add missing token usage span attributes during model usage (77bf325)
• Add param support to Bigtable execute_sql (5702a4b)
• Add RunSkillScriptTool to SkillToolset (636f68f)
• Add streaming support for Anthropic models (5770cd3), closes #3250
• Add support for ADK tools in SkillToolset (44a5e6b)
• Add support for MCP App UI widgets in MCPTool (86db35c)
• Add support for PDF documents in Anthropic LLM (4c8ba74)
• Add support for toolsets to additional_tools field of SkillToolset (066fcec)
• Add UiWidget to EventActions for supporting new experimental UI Widgets feature (530ff06)
• Added tracking headers for ADK CLI command to Agent Engine (3117446)
• auth: Add pluggable support for auth integrations using AuthProviderRegistry within CredentialManager (d004074)
• bigtable: add Bigtable cluster metadata tools (34c560e)
• durable runtime support (07fdd23)
• Enable output schema with tools for LiteLlm models (89df5fc), closes #3969
• Enhance BigQuery plugin schema upgrades and error reporting (bcf38fa)
• Enhance BQ plugin with fork safety, auto views, and trace continuity (80c5a24)
• execute-type param addition in GkeCodeExecutor (9c45166)
• Handle Conflict Errors in BigQuery Agent Analytics Plugin (372c76b)
• Introduce A2A request interceptors in RemoteA2aAgent (6f772d2)
• limit number of user-provided BigQuery job labels and reserve internal prefixes (8c4ff74)
• New implementation of A2aAgentExecutor and A2A-ADK conversion (87ffc55)
• New implementation of RemoteA2aAgent and A2A-ADK conversion (6770e41)
• otel: add gen_ai.tool.definitions to experimental semconv (4dd4d5e)
• otel: add experimental semantic convention and emit gen_ai.client.inference.operation.details event (19718e9)
• runners: pass GetSessionConfig through Runner to session service (eff724a)
• skill: Add BashTool (8a31612)
• Support all types.SchemaUnion as output_schema in LLM Agent (63f450e)
• telemetry: add new gen_ai.agent.version span attribute (ffe97ec)
• telemetry: add new gen_ai.agent.version span attribute (a61ccf3)
• telemetry: capture tool execution error code in OpenTelemetry spans (e0a6c6d)
• tools: add preserve_property_names option to OpenAPIToolset (078b516)
• Update docstring for BigQuery search tool (52b2f27)
• Warn when accessing DEFAULT_SKILL_SYSTEM_INSTRUCTION (35366f4)

Bug Fixes

• Add a script to the sample skills agent (8ad8bc9)
• Add usage field to ModelResponse in LiteLLM tests (245b2b9)
• Allow artifact services to accept dictionary representations of types.Part (b004da5), closes #2886
• Change Mypy workflow trigger to manual dispatch (d0825d8)
• Decode image data from ComputerUse tool response into image blobs (d7cfd8f)
• Expand LiteLLM reasoning extraction to include 'reasoning' field (9468487), closes #3694
• filter non-agent directoris from list_agents() (3b5937f)
• Fix Type Error by initializing user_content as a Content object (2addf6b)
• Handle length finish reason in LiteLLM responses (4c6096b), closes #4482
• In SaveFilesAsArtifactsPlugin, write the artifact delta to state then event actions so that the plugin works with ADK Web UI's artifacts panel (d6f31be)
• Make invocation_context optional in convert_event_to_a2a_message (8e79a12)
• Optimize row-level locking in append_event (d61846f), closes #4655
• Preserve thought_signature in FunctionCall conversions between GenAI and A2A (f9c104f)
• Preserve thought_signature in LiteLLM tool calls (ae565be), closes #4650
• Prevent splitting of SSE events with artifactDelta for function resume requests (6a929af), closes #4487
• Propagate file names during A2A to/from Genai Part conversion (f324fa2)
• Propagate thought from A2A TextPart metadata to GenAI Part (e59929e)
• Re-export DEFAULT_SKILL_SYSTEM_INSTRUCTION to skills and skill/prompt.py to avoid breaking current users (de4dee8)
• Refactor type string update in Anthropic tool param conversion (ab4b736)
• simulation: handle NoneType generated_content (9d15517)
• Store and retrieve EventCompaction via custom_metadata in Vertex AISessionService (2e434ca), closes #3465
• Support before_tool_callback and after_tool_callback in Live mode (c36a708), closes #4704
• temp-scoped state now visible to subsequent agents in same invocation (2780ae2)
• tools: Handle JSON Schema boolean schemas in Gemini schema conversion (3256a67)
• typo in A2A EXPERIMENTAL warning (eb55eb7)
• Update agent_engine_sandbox_code_executor in ADK (dff4c44)
• update Bigtable query tools to async functions (72f3e7e)
• Update expected UsageMetadataChunk in LiteLLM tests (dd0851a), closes #4680
• update toolbox server and SDK package versions (2e370ea)
• Validate session before streaming instead of eagerly advancing the runner generator (ebbc114)

Code Refactoring

• extract reusable functions from hitl and auth preprocessor (c59afc2)
• Rename base classes and TypeVars in optimization data types (9154ef5)
• use McpTool instead of MCPTool to resolve deprecation warnings (175d638)
• use McpTool instead of MCPTool to resolve deprecation warnings (af1c318)

---

This PR was generated with Release Please. See documentation.

[gemini-code-assist]

Summary of Changes

Hello, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request marks the release of version 1.27.0 of the ADK (Agent Development Kit). It incorporates a substantial collection of new features, critical bug fixes, and targeted code refactorings. The changes aim to expand the framework's capabilities, improve its stability, and enhance developer experience through better tooling, broader model support, and refined integration with various Google Cloud services like BigQuery and Bigtable.

Highlights

• New Features: Introduced a new adk optimize command, added streaming support for Anthropic models, and enhanced BigQuery plugin capabilities with fork safety, auto views, and trace continuity. Also, a new BashTool was added to the skill toolset.
• Improved Integrations and Tooling: Added pluggable authentication support via AuthProviderRegistry, introduced Bigtable cluster metadata tools, and enabled output schema with tools for LiteLLM models. Support for GCS filesystem for Skills (text and PDF) and ADK tools in SkillToolset was also added.
• Bug Fixes and Stability: Resolved issues such as allowing artifact services to accept dictionary representations of types.Part, handling length finish reasons in LiteLLM responses, and optimizing row-level locking in event appending. Various fixes related to A2A conversion and thought propagation were also implemented.
• Code Refactoring: Refactored base classes and TypeVars in optimization data types and updated usage of MCPTool to McpTool to address deprecation warnings.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

Changelog

• CHANGELOG.md
  • Added a new section for version 1.27.0, detailing a wide array of new features, bug fixes, and code refactorings.
  • New features include enhancements to ADK commands, model streaming support, BigQuery and Bigtable tool integrations, and OpenTelemetry semantic conventions.
  • Bug fixes address issues related to artifact services, LiteLLM responses, event handling, A2A conversions, and various other stability improvements.
  • Code refactorings involved extracting reusable functions and resolving deprecation warnings for tool classes.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for GitHub and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request automates the release for version 1.27.0. The changes include updating the version in the manifest and version file, and adding the new release notes to the changelog. My review identifies an incorrect future date in the CHANGELOG.md file that should be corrected.

[gemini-code-assist]

The release date (2026-03-11) is set in the future. This should be corrected to the actual release date. This appears to be a recurring issue in the changelog, as previous versions also have future dates. If the exact date is not known at generation time, consider using a placeholder like (YYYY-MM-DD) which can be replaced with the correct date upon release.

Suggested change

	## [1.27.0](https://github.com/google/adk-python/compare/v1.26.0...v1.27.0) (2026-03-11)
	## [1.27.0](https://github.com/google/adk-python/compare/v1.26.0...v1.27.0) (YYYY-MM-DD)