Bug Review

This is a heavyweight PR review system that runs five parallel analysis passes with shuffled diffs to reduce ordering bias, then uses majority voting to filter findings. A separate Claude Opus instance validates results without seeing the original passes. It posts inline GitHub comments and can generate autofix commits with test validation and scope checks. The interesting part is the learning loop: after PRs merge, you run a resolve command to classify whether findings were actually fixed, then the system adjusts category weights to suppress noisy issue types over time. Requires gh CLI and assumes you want a thorough, opinionated review process rather than a quick scan. Overkill for small changes, but if you're serious about catching bugs in high-stakes PRs and willing to let it learn from your merge decisions, it's built for that.

skills.sh installs: 256
GitHub stars: 222
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add pproenca/dot-skills --skill bug-review --agent claude-code

Installs into .claude/skills of the current project.

Files

Bug Review v2

Multi-pass PR review agent with 5 parallel review passes, majority voting, independent Opus validation, and resolution rate learning. Posts inline PR comments and optionally generates autofix commits. Tracks whether findings get resolved at merge time and uses that signal to improve future reviews.

When to Apply

User asks to review a pull request for bugs or correctness issues
User runs /bug-review <PR-number-or-URL>
User runs /bug-review:resolve <PR> to classify resolutions after merge
User runs /bug-review:report for resolution rate statistics
User asks for code review focused on logic errors, edge cases, or security
User wants to find bugs in a diff or set of changes

Setup

On first run, verify:

gh CLI is installed and authenticated (gh auth status)
Current directory is a git repo with a GitHub remote
jq is installed (for JSON processing)
bc is installed (for resolution rate calculations; pre-installed on most systems)

Read config.json for configuration (passes, vote threshold, models, category weights).