code-review

Analyze code changes for correctness, security, and design issues.

1|Updated Mar 10, 2026
One-click install
npx skills add https://github.com/butterflyskies/claude-skills --skill code-review-butterflyskies
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-review
Source: https://github.com/butterflyskies/claude-skills/tree/main/code-review
Command: npx skills add https://github.com/butterflyskies/claude-skills --skill code-review-butterflyskies

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill automates the process of code review, identifying critical bugs, design flaws, and security vulnerabilities before they impact production.

Core Features & Use Cases

  • Multi-Agent Analysis: Employs specialized AI agents to analyze code from different perspectives (correctness, design, security).
  • Contextual Understanding: Gathers project conventions, architecture, and caller information for thorough reviews.
  • Actionable Findings: Provides concrete fixes for identified issues, categorized by severity (P1, P2, P3).
  • Use Case: Automatically review a pull request, ensuring adherence to best practices and identifying potential regressions.

Quick Start

Use the code-review skill to review the current branch's changes against its base.

Frequently Asked Questions about code-review

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

FAQPage Schema
How do I automate code review for a pull request to find security vulnerabilities?

Automated code review analyzes pull requests using specialized AI agents to evaluate correctness, design, and security. It identifies vulnerabilities and design flaws before production impact, providing actionable findings with concrete fixes and severity levels.

Can I review staged changes or specific files instead of a full branch?

Code review supports various scopes including staged changes, individual files, branches, and pull requests. The multi-agent approach systematically analyzes the targeted scope, integrating project conventions and caller information for thorough contextual evaluation.

How does AI code review detect bugs and maintainability issues?

AI code review employs specialized agents to analyze code from different perspectives including correctness, safety, design, and maintainability. It gathers project architecture and convention context to identify potential regressions and provide severity-categorized actionable findings.

Does automated code review work with existing project conventions and architecture?

Automated code review integrates with existing project conventions, architecture, and symbol information. This contextual understanding enables comprehensive analysis of caller relationships and design patterns, ensuring reviews align with established codebase standards.

What's the best way to ensure code quality before merging changes?

Reviewing the current branch's changes against its base using multi-agent analysis ensures adherence to best practices. The review identifies critical bugs and security vulnerabilities with concrete fixes, categorized by P1, P2, and P3 severity levels for prioritized resolution.

What limitations should I expect from AI-assisted code review?

AI-assisted code review relies on gathered project context, conventions, and symbol information for analysis accuracy. While it identifies bugs, design flaws, and security vulnerabilities across multiple dimensions, findings require human validation before applying suggested concrete fixes.