What problem does it solve?
PR reviewers and teams spend significant time gathering PR metadata, diffs, comments, and related issues, then manually producing structured review documents and navigating a two-stage approval workflow. This Skill automates PR data collection, organizes a review workspace, analyzes changes against industry-standard criteria, and generates internal and public review artifacts with an optional inline-comment workflow.
Core Features & Use Cases
- Automated Data Collection: Fetches PR metadata, diffs, comments, commits, and related issues via GitHub CLI
- Systematic Analysis: Applies a comprehensive quality checklist against the PR changes
- Structured Review Files: Generates internal review, clean public review, and inline comment templates
- Two-Stage Approval: Nothing posts to GitHub until explicit approval
- Inline Comments: Optionally adds code-specific feedback directly to the PR
- Ticket Tracking: Extracts and links JIRA/GitHub issue references
- Templates & Workflow: Provides professional templates and commands to streamline posting
Quick Start
Use this skill to fetch a PR, analyze it, and generate review artifacts:
- Run data collection: python scripts/fetch_pr_data.py <pr_url>
- Review generated artifacts in the workspace (e.g., /tmp/PRs/<repo>/<pr>/)
- Generate review files: python scripts/generate_review_files.py /tmp/PRs/<repo>/<pr> --findings findings.json
- Open the review directory for editing: /show
- Post and approve when ready: /send or request changes: /send-decline