-
Notifications
You must be signed in to change notification settings - Fork 2
Description
Describe the bug
First of all, thank you for providing such a helpful tool.
Hello,
I would like to report an issue with Amazon Q Agent.
When the agent creates a pull request and the CI checks fail, it still posts the following message:
✅ I finished the code review, and didn't find any security or code quality issues..
This is misleading, as the agent itself wrote the code and the CI has already failed.
In this case, the agent should not mark the PR as reviewed or complete.
Instead, it should recognize the CI failure and request fixing the issues it caused before considering the PR as reviewed.
Could you please investigate and address this issue?
Thank you in advance.
Expected Behavior
When the agent creates a pull request and the CI checks fail, it should not consider the code review complete.
Instead, the agent should automatically fix the issues until the CI passes successfully, and only then mark the pull request as reviewed and complete.
Current Behavior
When the agent creates a pull request and the CI checks fail, it still posts the following message:
✅ I finished the code review, and didn't find any security or code quality issues..
Reproduction Steps
-
Use Amazon Q Agent to generate code and create a pull request.
-
Ensure that the generated code has issues that will cause CI to fail (for example, intentionally break coding style or introduce a simple syntax error).
-
Wait for the CI to run and fail.
-
Observe that despite the CI failure, the agent posts the following message:
✅ I finished the code review, and didn't find any security or code quality issues..
This completes the pull request review incorrectly even though CI did not pass.
Possible Solution
No response
Additional Information/Context
No response