All messages converted to thoughts in a2a

[mfojtak]

Describe the bug
All messages are converted to thoughts when using a2a remote agent.

This is because:

if event.content and update.status.state in [
              TaskState.submitted,
              TaskState.working,
          ]:
            for part in event.content.parts:
              part.thought = True

code in agents/remote_a2a_agent.py

This is not appropriate as it converts genuine content messages into thoughs. Removing this code fixes the problem.

To Reproduce
Use remote a2a connection in version 1.20 or 1.19

Expected behavior
Nature of a2a should be respected and not blindly converting all of them to thoughts.

Screenshots
If applicable, add screenshots to help explain your problem.

Desktop (please complete the following information):

• OS: [e.g. macOS, Linux, Windows] Linux
• Python version(python -V): 3.13
• ADK version(pip show google-adk): 1.20

Model Information:

• Are you using LiteLLM: Yes
• Which model is being used(e.g. gemini-2.5-pro) - Bedrock

Additional context
Add any other context about the problem here.

[adk-bot]

Notification: The author has updated the issue description. Maintainers, please review.

[adk-bot]

This issue has been automatically marked as stale because it has not had recent activity for 7 days after a maintainer requested clarification. It will be closed if no further activity occurs within 7 days.

[adk-bot]

This has been automatically closed because it has been marked as stale for over 7 days.

[mfojtak]

The problem still persists in 1.22.1. All a2a messages are treated as thoughts

[surajksharma07]

@mfojtak The root cause is in remote_a2a_agent.py:435-462 — the _handle_a2a_response method blindly sets part.thought = True for all content when task state is submitted or working, which hides genuine agent responses.

As a workaround, you can subclass RemoteA2aAgent and override _handle_a2a_response to skip the two thought-marking loops (lines 446-447 and 461-462).

This bug is still present in the latest source.

[surajksharma07]

@seanzhou1023 Please have a look into it once.