What problem does it solve? Manually reviewing code changes for bugs, security vulnerabilities, and quality issues is slow and error-prone. This Skill automates the review cycle by running CodeRabbit against your staged, committed, or branch-compared changes and turning findings into an actionable task list. ## Core Features & Use Cases - Automated Review Execution: Runs coderabbit review --agent on all, committed, or uncommitted changes, with support for base branch, base commit, and directory scoping. - Severity-Grouped Findings: Organizes results into Critical, Warning, and Info tiers so you fix the highest-risk issues first. - Autonomous Fix Loop: Implements a feature, reviews it, fixes critical and warning issues, and re-reviews until only info-level findings remain. - Use Case: After finishing a new API endpoint, ask for a review against the main branch; the Skill runs CodeRabbit, lists the findings by severity, and iteratively fixes them. ## Quick Start Review my uncommitted changes with CodeRabbit and fix any critical or warning issues it finds.