code-review

Review local git changes and PR diffs for correctness, security, design, testing, and documentation.

3|Updated Mar 4, 2026
One-click install
npx skills add https://github.com/scute-sh/scute --skill code-review-scute-sh
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: code-review
Source: https://github.com/scute-sh/scute/tree/main/.claude/skills/code-review
Command: npx skills add https://github.com/scute-sh/scute --skill code-review-scute-sh

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill automates the process of reviewing code for correctness, security vulnerabilities, design quality, testing coverage, and documentation completeness, ensuring higher code quality and reducing manual review burden.

Core Features & Use Cases

  • Multi-faceted Review: Employs three specialized sub-agents to cover correctness/security, design/quality, and testing/documentation.
  • Contextual Analysis: Reads entire files and neighboring code to understand changes in context.
  • Structured Findings: Generates detailed, severity-ranked findings in markdown files.
  • Use Case: Automatically review a pull request to identify potential bugs, security flaws, and areas for code improvement before merging.

Quick Start

Use the code-review skill to review all local git changes.

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?

You can automate code review for a pull request by analyzing local git changes and PR diffs. The multi-agent approach evaluates correctness, security, design, testing, and documentation, generating structured findings categorized by severity.

What does automated code quality analysis check for?

Automated code quality analysis checks for correctness, security vulnerabilities, design flaws, testing coverage, and documentation completeness. It reads entire files and neighboring code to evaluate changes within their full context.

How are code review findings structured and reported?

Code review findings are structured as detailed markdown files categorized by severity levels: CRITICAL, MAJOR, and MINOR. This ranking helps identify potential bugs, security flaws, and areas for code improvement.

Can I review local git changes before committing?

Yes, you can review local git changes before committing. The analysis gathers file manifests from local changes, commits, or PR diffs to perform comprehensive quality and correctness checks.

How does multi-agent code review improve maintainability?

Multi-agent code review improves maintainability by employing specialized sub-agents for correctness/security, design/quality, and testing/documentation. This contextual analysis identifies potential issues in code quality across multiple dimensions.