code-review

Automate PR code reviews with parallel multi-agent analysis and AGENTS.md compliance checks.

5|Updated Dec 28, 2025
One-click install
npx skills add https://github.com/Bind/skillz.sh --skill code-review-bind
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-review
Source: https://github.com/Bind/skillz.sh/tree/main/docs/registry/skills/code-review
Command: npx skills add https://github.com/Bind/skillz.sh --skill code-review-bind

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill automates PR code reviews using a multi-agent analysis framework to ensure consistency, speed, and accountability in code changes.

Core Features & Use Cases

  • AGENTS.md compliance auditing: Audits PR changes against repository guidelines and cites exact AGENTS.md rules when violations occur.
  • Bug and logic detection: Scans diffs for obvious bugs and potential logic errors introduced in the pull request.
  • Collaborative review workflow: Runs four specialized agents in parallel to surface high-signal issues for faster triage.
  • Use Case: A large, multi-file PR is automatically analyzed and issues are surfaced with precise file and line references for quick remediation.

Quick Start

Run the automated code-review workflow:

  1. Check readiness: bun .opencode/skill/github-pr/check-review-needed.js
  2. Gather guidelines: bun .opencode/skill/github-pr/list-guideline-files.js --json
  3. Summarize PR: gh pr view <PR-NUMBER> && gh pr diff <PR-NUMBER>

Frequently Asked Questions about code-review

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate PR code reviews for multi-file GitHub pull requests?

Automate PR code reviews by running a multi-agent analysis pipeline that targets repository diffs to provide structured, high-signal feedback on changes. Four specialized agents execute in parallel to surface bugs, logic issues, and AGENTS.md guideline violations.

How does multi-agent code review work for detecting bugs in a pull request?

Multi-agent code review works by running four specialized agents in parallel to scan PR diffs for obvious bugs and potential logic errors. This parallelized pipeline surfaces high-signal issues with precise file and line references for quick remediation.

Can I check GitHub PR changes against AGENTS.md rules automatically?

Yes, you can check GitHub PR changes against AGENTS.md rules automatically. The review framework audits pull request diffs for compliance and cites exact AGENTS.md rules when violations occur, ensuring accountability in code changes.

Do I need GitHub CLI and Bun to run automated code review workflows?

You need GitHub CLI and Bun to run automated code review workflows. The quick start process uses Bun to execute readiness checks and gather guideline files, while GitHub CLI summarizes PR views and diffs for the multi-agent analysis.

What is the best way to structure code review feedback for large pull requests?

The best way to structure code review feedback for large pull requests is using a parallelized multi-agent framework. It isolates bug detection, logic analysis, and AGENTS.md compliance auditing to generate clear provenance in issue reporting.

What are the limitations of automated multi-agent code review for complex logic?

Automated multi-agent code review for complex logic primarily targets obvious bugs and potential errors introduced in the pull request diff. It focuses on high-signal issue surfacing and AGENTS.md compliance rather than deep architectural analysis.