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
- Support max_parallel_workers on Agent parallel workers
- RFC: supervised task handles, queued steering, and explicit result merges
- feat: add owned parallel task collaboration and selective cancellation
- Evaluation judge and user simulator default to gemini-2.5-flash, which is no longer available to new users
- fix(evaluation): isolate stored eval sets from caller mutations
- ModelArmorPlugin fails with "Event loop is closed" after one sync Runner.run() call: its client is not cached per event loop
- fix(models): keep streamed ApigeeLlm tool call fragments without an index together
- ApigeeLlm raises ValueError instead of reporting MAX_TOKENS when streamed tool arguments are cut off
- After a tool confirmation resumes a `Workflow`, an `LlmAgent` node's model receives the workflow's user message instead of the node's input
- fix(evaluation): don't score a judge response that leaves a rubric without a verdict
- Docs
- Python not yet supported