adk-python
https://github.com/google/adk-python
Python
An open-source, code-first Python toolkit for building, evaluating, and deploying sophisticated AI agents with flexibility and control.
Triage Issues!
When you volunteer to triage issues, you'll receive an email each day with a link to an open issue that needs help in this project. You'll also receive instructions on how to triage issues.
Triage Docs!
Receive a documented method or class from your favorite GitHub repos in your inbox every day. If you're really pro, receive undocumented methods or classes and supercharge your commit history.
Python not yet supported2 Subscribers
Add a CodeTriage badge to adk-python
Help out
- Issues
- fix: filter hidden artifacts from LLM context
- A custom BaseLlm has no documented mapping from LlmRequest.config to a Vertex request body, and getting it wrong drops your tools with no error
- Hide certain artifacts from the LLM in LoadArtifactsTool
- to_openeval()/from_openeval(): interop between EvalSet and the EvalPort portable format
- fix(events): serialize deferred Pydantic state models
- feat(cli): add `adk eval --num_runs N` with per-eval-case aggregation
- Support the MCP Tasks extension (SEP-2663) in McpToolset: expose the mcp 2.x client extension seam
- MCPSessionManager pools a server-terminated streamable-HTTP session forever: `_is_session_disconnected` only checks local stream flags (scope of closed #3321, still broken in v2.7.1)
- Support custom header names for MCP servers secured with Auth Providers
- fix(a2a): map Event.output into to_a2a artifacts
- Docs
- Python not yet supported