What problem does it solve? Pull request feedback arrives scattered across review comments, bot reviews, CI failures, and merge conflicts, making it easy to drop findings or ship unreviewed changes. This Skill builds a complete feedback ledger, verifies every claim against source code, fixes confirmed issues, and reports closure status for every item. ## Core Features & Use Cases - Complete Feedback Ledger: Normalizes every review comment, CI failure, conflict, and bot finding into tracked FB-### items so nothing is silently dropped. - Skeptical Claim Verification: Classifies each item as CONFIRMED, DISPROVED, PRE_EXISTING, or NEEDS_USER_DECISION using source-level evidence, including dedicated protocols for bot security findings. - Multi-Round Bot Review Handling: Continues the ledger across iterative auto-review bot rounds with a 3-strikes evidence-escalation rule instead of re-triaging from scratch. - Mandatory Gate Discipline: Enforces structural pre-checks, reviewer and test gates, and a closeout gate before any fix lands or the PR is published. - Use Case: A reviewer bot posts new findings after every push to your PR. Use this Skill to triage each round, disprove false positives with code evidence, fix real issues, and post a per-round closure ledger comment. ## Quick Start Ask the agent to resolve all feedback on the current pull request using the swarm-pr-feedback workflow, verifying each review comment and CI failure before fixing.
