Skip to main content
skill CX Codex

code-review-analysis — Claude Code skill: structured code review

A skill that walks the agent through a systematic review: code quality, security, performance, tests. Useful when you are checking a pull request before merge.

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add aj-geddes/useful-ai-prompts --skill code-review-analysis -a claude-code
Open on GitHub

What code-review-analysis does

The skill sets an order for reading changes. The agent starts with diffs and commit history — git diff main...feature-branch, git diff --stat, git log --oneline — then works through checklists: code quality, security, performance, testing, standards compliance.

Under the hood there is a set of guides in the references/ directory: initial assessment, code quality analysis, security review, performance review, testing review, and a separate best-practices file. The agent pulls the section that fits the task instead of reading everything.

It also carries reviewer etiquette: explain the reasoning behind a suggestion, show code examples, ask when something is unclear, call out good practices. The don'ts are spelled out too — no nitpicking style, no blocking on subjective preferences, no reviewing more than 400 lines at once. That last one is worth remembering: on a large diff the review goes fuzzy, so splitting the check into chunks is the sensible move.

What it helps with

  • review a pull request before merging
  • spot security holes in the changed code
  • check whether new branches are covered by tests
  • write feedback with code examples for the author
  • verify the code against project standards

Example prompts for the agent

  • review the feature-branch using the code-review-analysis skill
  • check the main...feature-branch diff for security and test coverage
  • go through this PR and explain each comment you leave

How to install code-review-analysis

1Skills CLI (Vercel)

npx skills add aj-geddes/useful-ai-prompts --skill code-review-analysis -a claude-code

2Manually: put the files in place

~/.claude/skills/code-review-analysis/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/code-review-analysis/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about code-review-analysis

How do I install code-review-analysis in Claude?

Run npx skills add aj-geddes/useful-ai-prompts --skill code-review-analysis -a claude-code. The skill lands in ~/.claude/skills/code-review-analysis/SKILL.md.

What does code-review-analysis do?

It runs changes through checklists for code quality, security, performance, tests and standards. Each comment comes with reasoning and code examples.

Which platforms can run this skill?

Besides Claude there are install paths for Codex, Cursor, GitHub Copilot, Gemini CLI and OpenCode — the commands differ by the -a flag or a separate installer.

Source text

SKILL.md · MIT
---
name: code-review-analysis
description: >
  Perform comprehensive code reviews with best practices, security checks, and
  constructive feedback. Use when reviewing pull requests, analyzing code
  quality, checking for security vulnerabilities, or providing code improvement
  suggestions.
---

# Code Review Analysis

## Table of Contents

- [Overview](#overview)
- [When to Use](#when-to-use)
- [Quick Start](#quick-start)
- [Reference Guides](#reference-guides)
- [Best Practices](#best-practices)

## Overview

Systematic code review process covering code quality, security, performance, maintainability, and best practices following industry standards.

## When to Use

- Reviewing pull requests and merge requests
- Analyzing code quality before merging
- Identifying security vulnerabilities
- Providing constructive feedback to developers
- Ensuring coding standards compliance
- Mentoring through code review

## Quick Start

Minimal working example:

```bash
# Check the changes
git diff main...feature-branch

# Review file changes
git diff --stat main...feature-branch

# Check commit history
git log main...feature-branch --oneline
```

## Reference Guides

Detailed implementations in the `references/` directory:

| Guide | Contents |
|---|---|
| [Initial Assessment](references/initial-assessment.md) | Initial Assessment |
| [Code Quality Analysis](references/code-quality-analysis.md) | Code Quality Analysis |
| [Security Review](references/security-review.md) | Security Review |
| [Performance Review](references/performance-review.md) | Performance Review |
| [Testing Review](references/testing-review.md) | Testing Review |
| [Best Practices](references/best-practices.md) | Best Practices |

## Best Practices

### ✅ DO

- Be constructive and respectful
- Explain the "why" behind suggestions
- Provide code examples
- Ask questions if unclear
- Acknowledge good practices
- Focus on important issues
- Consider the context
- Offer to pair program on complex issues

### ❌ DON'T

- Be overly critical or personal
- Nitpick minor style issues (use automated tools)
- Block on subjective preferences
- Review too many changes at once (>400 lines)
- Forget to check tests
- Ignore security implications
- Rush the review

© aj-geddes · MIT · aj-geddes/useful-ai-prompts

Similar skills

Everything for Code review and refactoring →

From the same set118

Project sponsors

Companies that keep this analytics open
Gold sponsor

Impressions, clicks and code copies — open statistics. We keep the numbers public for the same reason NeBlask itself is public. Become a sponsor