review-retrospective

To prevent recurring code review failures, this skill analyzes maintainer feedback, clusters systemic review misses, and dispatches automated jobs to update pre-push gates. It closes the feedback loop by ensuring missed edge cases are durably prevented rather than just fixed individually.

Double-loop learning for pull request feedback treats maintainer comments as indictments of the review process itself. It automatically clusters recurring misses to identify systemic patterns, updating the agent's review rubric to prevent similar errors from reaching a human again.

Automating process improvement involves analyzing maintainer feedback to identify review process failures and dispatching improvement jobs. The skill updates juror-seat briefs with deterministic record-keeping to ensure continuous improvement of your agentic workflow.

Yes, you can use automated feedback loops to update pre-push gates. When maintainer feedback identifies a missed edge case and clusters it with similar failures to meet a threshold, the skill automatically implements durable pre-push checks to catch the issue.

You need to update juror-seat briefs when maintainer feedback reveals systemic review process failures. By clustering recurring misses and meeting a threshold, the skill automatically updates these briefs to prevent identical edge cases from bypassing the review gauntlet again.

The best way to cluster recurring code review misses is treating maintainer feedback as process indictments rather than individual code symptoms. This approach identifies systemic patterns and dispatches automated improvement jobs to update review rubrics and pre-push gates.