sometimes the AI code has failed CI checks like linter error in PR: apache/casbin#1691
we need to monitor such failed CI checks for each PR in the github repos of the monitored github orgs
if the CI failed, then reply in the PR and "AT" the copilot and provide the details, so let copilot fix it
it needs to monitor the fix process. if we found this bug is fixed, then next error happens, then we fix the next error
if the same error cannot be fixed by AI like for 3 times, then AI should stop, or it will run endlessly without progress.
for all human-created PR, it should automatically let copilot review the code
sometimes the AI code has failed CI checks like linter error in PR: apache/casbin#1691
we need to monitor such failed CI checks for each PR in the github repos of the monitored github orgs
if the CI failed, then reply in the PR and "AT" the copilot and provide the details, so let copilot fix it
it needs to monitor the fix process. if we found this bug is fixed, then next error happens, then we fix the next error
if the same error cannot be fixed by AI like for 3 times, then AI should stop, or it will run endlessly without progress.
for all human-created PR, it should automatically let copilot review the code