fix-pr

To fix a failing pull request from CI logs, this skill diagnoses log-derived failure types like formatting, linting, build, or test failures, and applies the corresponding repository remediation steps automatically. It then rebases and pushes the updated commit.

CI failure identification works by delegating log analysis to a dedicated subagent that classifies failures into categories such as format, lint, build, unit tests, or e2e, surfacing actionable details for targeted test remediation.

Yes, you can use this to rebase and fix formatting failures in AztecProtocol packages. It validates the PR via gh, checks out and rebases onto the base branch without merges, and runs the correct yarn-workspace commands to resolve formatting issues.

The best way to resolve unit test or build failures in a yarn workspace PR is to run a targeted fix loop that executes the appropriate package-scoped yarn commands, iterating until the quality gates pass, and then amends and pushes the commit.

This PR remediation workflow uses a rebase-based approach instead of merge commits to keep the repository history clean and avoid merge conflicts when checking out and updating the failing pull request.

Limitations of automated CI log debugging include its intended scope for AztecProtocol/aztec-packages PRs specifically, requiring PR validation via gh, and dependency on a rebase-based workflow without merge commits to push the amended changes.