What problem does it solve?
Managing pull request reviews across large or complex changes requires coordinating multiple reviewers, tests, and validations manually. This Skill automates PR lifecycle management by spawning AI agent swarms that analyze diffs, run reviews, and coordinate merge decisions directly from GitHub pull requests.
Core Features & Use Cases
- PR-Based Swarm Creation: Initialize agent swarms from PR metadata, diffs, and labels using the gh CLI and ruv-swarm commands, with topology selected by PR size.
- Automated Code Review: Spawn security, performance, and style review agents that post comments and progress updates back to the PR.
- Label-Driven Agent Assignment: Map PR labels like bug, feature, or refactor to appropriate agent types for targeted review.
- Use Case: A large feature PR over 500 lines is opened; the Skill initializes a hierarchical swarm with architect, coder, tester, and security agents, posts review comments via gh CLI, and enables auto-merge once swarm validation passes.
Quick Start
Ask the agent to initialize a swarm for pull request 123, analyze its diff, run a multi-agent review, and post the results as PR comments.