Code Review Skill

When to Use

Activate when the user wants to:

• •Review a pull request or set of code changes
• •Get a multi-perspective analysis of code quality
• •Audit code for security, performance, or architecture concerns
• •Compare branches or review specific files

User Preferences

Before reviewing, resolve the user's Profile:

1. •Check for .github/copilot-preferences.yml or .vscode/copilot-preferences.yml in the workspace.
2. •Apply any custom review conventions, severity thresholds, or focus areas.
3. •For the full resolution order (including policy packs), see the user-preferences skill.

Workflow

1. Understand the Request

Determine what needs to be reviewed:

• •Pull Request: Analyze a specific PR by number or URL
• •Current Changes: Review uncommitted changes in the workspace
• •File or Directory: Review specific files or directories
• •Branch Comparison: Compare two branches

2. Gather Context

• •Fetch PR details if reviewing a pull request
• •Read modified files and understand changes
• •Detect project type (Rails, TypeScript, Python, etc.)
• •Review related code and dependencies

3. Detect Project Type

Identify the project type to tailor the review:

• •Rails: Gemfile with rails gem, config/application.rb, app/ structure
• •TypeScript: tsconfig.json, .ts/.tsx files
• •Python: requirements.txt/pyproject.toml, .py files

4. Multi-Agent Analysis

Coordinate specialized analysis perspectives:

Always Engaged:

• •Architecture analysis — system design and architectural patterns
• •Security audit — vulnerabilities and security risks
• •Performance review — characteristics and optimization
• •Code simplicity — clarity and simplification opportunities
• •Pattern analysis — design patterns and consistency
• •Data integrity — database operations and data integrity

Language-Specific (invoke when applicable):

• •Rails projects: Rails reviewer and DHH philosophy reviewer
• •TypeScript projects: TypeScript reviewer
• •Python projects: Python reviewer

5. Review Perspectives

Technical Excellence: Code quality, best practices, test coverage, documentation Security: Input validation, auth, data protection, OWASP Top 10 Performance: Algorithmic complexity, query optimization, caching, scalability Architecture: Component boundaries, pattern consistency, dependency management Maintainability: Readability, simplicity, error handling, observability

6. Categorize Findings

Severity Levels:

• •🔴 CRITICAL (P1): Security vulnerabilities, data loss risks, breaking changes
• •🟡 IMPORTANT (P2): Performance issues, architectural concerns, significant bugs
• •🔵 NICE-TO-HAVE (P3): Code quality improvements, minor optimizations

7. Output Format

# Code Review Summary

## Overview
[Brief summary of what was reviewed]

## Project Type
[Rails/TypeScript/Python/etc.]

## Key Findings

### 🔴 Critical Issues (P1)
1. **[Issue Title]** - `file:line`
   - Problem: [Description]
   - Impact: [Why critical]
   - Solution: [How to fix]

### 🟡 Important Issues (P2)
...

### 🔵 Suggestions (P3)
...

## Agent Insights
[Summaries from each analysis perspective]

## Overall Assessment
**Risk Level**: [Low/Medium/High]
**Recommended Action**: [Approve / Approve with changes / Request changes]

## Next Steps
1. [Immediate action items]
2. [Follow-up tasks]

Guardrails

• •Be specific: reference exact file paths and line numbers.
• •Be constructive: focus on improvements, not criticism.
• •Prioritize: highlight the most important issues first.
• •Suggest solutions: don't just identify problems, propose fixes.