What problem does it solve? Addressing pull request review feedback is tedious: you must read every comment, decide which suggestions are valid, make code changes, reply to reviewers, and resolve threads one by one. This Skill automates that entire loop, triaging feedback and dispatching parallel agents to fix valid issues while declining harmful suggestions with cited reasons. ## Core Features & Use Cases - Automated feedback triage: Fetches unresolved review threads, PR comments, and review bodies via the GitHub GraphQL API, then separates new actionable items from already-handled or non-actionable bot noise. - Parallel fix agents: Spawns resolver agents for each thread, with cross-invocation cluster analysis that detects recurring feedback patterns across review rounds and batches work while avoiding file conflicts. - Full resolution loop: Validates the combined diff with the project's test suite, commits and pushes fixes, posts quoted replies, and resolves threads via GraphQL mutations. - Use Case: After reviewers leave 12 comments on your PR, run the Skill to have every valid issue fixed, tested, committed, replied to, and resolved, with a summary of verdicts and any items needing human decisions. ## Quick Start Ask the agent to resolve all review feedback on the current branch's pull request, or pass a PR number or specific comment URL to target.
