Skip to content

Conversation

@dbschmigelski
Copy link
Owner

Description

Related Issues

Documentation PR

Type of Change

Bug fix
New feature
Breaking change
Documentation update
Other (please describe):

Testing

How have you tested the change? Verify that the changes do not break functionality or introduce warnings in consuming repositories: agents-docs, agents-tools, agents-cli

  • I ran hatch run prepare

Checklist

  • I have read the CONTRIBUTING document
  • I have added any necessary tests that prove my fix is effective or my feature works
  • I have updated the documentation accordingly
  • I have added an appropriate example to the documentation to outline the feature, or no new docs are needed
  • My changes generate no new warnings
  • Any dependent changes have been merged and published

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.

zastrowm and others added 30 commits July 21, 2025 11:17
…#505)

Similar to strands-agents/pull/475 but using a dedicated github icon.

The github icon is the lite logo but copied/renamed to make it dedicated to github
These were necessary when we had conditional running but we switched to needing to approve all workflows for non-maintainers, so we no longer need these.

Co-authored-by: Mackenzie Zastrow <[email protected]>
…ectory (strands-agents#517)

- Add new section showcasing Agent(load_tools_from_directory=True) functionality
- Document automatic tool loading and reloading from ./tools/ directory
- Include practical code example for developers
- Improve discoverability of this development feature
Fixes issue strands-agents#50

Customers naturally want to pass nested collections of tools - the above issue has gathered enough data points proving that.
…ts#524)

* feat(a2a): support mounts for containerized deployments

* feat(a2a): escape hatch for load balancers which strip paths

* feat(a2a): formatting

---------

Co-authored-by: jer <[email protected]>
strands-agents#576)

* fix(event_loop): raise dedicated exception when encountering max tokens stop reason

* fix: update integ tests

* fix: rename exception message, add to exception, move earlier in cycle

* Update tests_integ/test_max_tokens_reached.py

Co-authored-by: Nick Clegg <[email protected]>

* Update tests_integ/test_max_tokens_reached.py

Co-authored-by: Nick Clegg <[email protected]>

* linting

---------

Co-authored-by: Nick Clegg <[email protected]>
)

Using a string causes the Anthropic API call to fail:
```
anthropic.BadRequestError: Error code: 400 - {'type': 'error', 'error': {'type': 'invalid_request_error', 'message': 'max_tokens: Input should be a valid integer'}}
```
…rands-agents#643)

Previously (strands-agents#642) bedrock would hang during message conversion because the exception was not being caught and thus the queue was always empty. Now all exceptions during conversion are caught

Co-authored-by: Mackenzie Zastrow <[email protected]>
dbschmigelski and others added 25 commits November 18, 2025 13:33
…andler (strands-agents#1211)

Make the verbose description and counting of tool use optional in PrintingCallbackHandler.
---------

Co-authored-by: Dean Schmigelski <[email protected]>
* fix: fix swarm session management integ test.

* share thread context (strands-agents#1146)

* async hooks (strands-agents#1119)

* fix: remove debug lines

---------

Co-authored-by: Patrick Gray <[email protected]>
…trands-agents#1231)

* fix(mcp): protect connection on non-fatal client side timeout error

* remove empty _MCP_CLIENT.md

* remove print statements

* remove test
Introduce bidirectional streaming capabilities to Strands SDK, enabling real-time voice and audio conversations with AI models through persistent streaming connections.

Bidirectional streaming moves beyond traditional request-response patterns by maintaining long-running conversations where users can interrupt, provide continuous input, and receive real-time audio responses. This implementation is marked as experimental as we refine the API based on user feedback and evolving model capabilities.

---------

Co-authored-by: Murat Kaan Meral <[email protected]>
Co-authored-by: Patrick Gray <[email protected]>
…manager (strands-agents#1274)

* Remove toolUse message when its missing due to pagination in session manager
strands-agents#1290)

* fix(agent): Return structured output JSON when AgentResult has no text

When AgentResult has no text content but structured_output is present,
__str__() now returns the JSON serialization of the structured output
instead of an empty string. This fixes output propagation in multi-agent
graphs where structured output was being lost.

Changes:
- Modified AgentResult.__str__() to fall back to structured_output.model_dump_json()
- Added unit test test__str__empty_message_with_structured_output to verify fix
- All existing tests pass, maintaining backward compatibility

strands-agents#1118
We keep getting warnings about odd indentation; this takes care of that

Co-authored-by: Mackenzie Zastrow <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

[BUG] A2A Doesn't Work on 1.2.0