Prefl AI - Your Dream Code Review Tool ✨

Professional AI-powered code review that catches bugs before they reach production

Website: prefl.run

🚀 Features

• 🔍 Comprehensive Analysis - Detects runtime errors, security vulnerabilities, memory leaks, performance issues
• 🎯 Smart Context Tracking - Analyzes related code and imports to maintain full context
• 🚨 Critical Issue Blocking - Prevents commits with critical problems
• 🌍 Multi-Language Support - Works with any programming language
• ⚡ Lightning Fast - Analyzes only staged changes, not entire codebase
• 🎨 Beautiful Output - Clear, emoji-rich reports with actionable fixes
• 💾 Export Reports - Save detailed analysis to files
• 🪝 Git Integration - Auto-runs on pre-commit hooks

📦 Installation

npm i -g @preflight-ai/cli@latest

🛠️ Setup

1. Initialize in your project

cd your-project
prefl init

This will:

• ✅ Create prefl.json config file
• ✅ Create .env and ask for your GROQ API key
• ✅ Add .env to .gitignore
• ✅ Install pre-commit hook automatically

2. Get your Groq API Key

1. Visit console.groq.com
2. Create a free account
3. Generate an API key
4. Paste it when prompted during prefl init

🎯 Usage

Automatic Review (Recommended)

Once initialized, Prefl automatically reviews your code before every commit:

git add .
git commit -m "feat: add new feature"
# ✨ Prefl analyzes your changes automatically!

Manual Analysis

# Analyze staged changes
prefl analyze

# Analyze entire repository
prefl analyze --all

# Save report to file
prefl analyze --output my-report.txt

# JSON output for CI/CD
prefl analyze --format json

Commands

prefl init                    # Setup Prefl in your project
prefl analyze                 # Analyze staged changes
prefl analyze --all           # Analyze entire repo
prefl analyze --output FILE   # Save results to file
prefl fix                     # Generate AI-suggested patches
prefl fix --apply             # Auto-apply generated fixes
prefl --version               # Show version
prefl --help                  # Show help

📊 What Gets Analyzed?

🚨 CRITICAL ISSUES (Blocks Commits)

• Runtime Errors: Null/undefined access, type mismatches, unhandled promises
• Security Vulnerabilities: SQL injection, XSS, CSRF, exposed secrets, insecure auth
• Memory Leaks: Event listeners without cleanup, circular references
• Data Loss: Missing validation, race conditions, improper async handling

⚠️ WARNINGS (Alerts)

• Performance: N+1 queries, unnecessary re-renders, blocking operations
• Code Smells: Tight coupling, magic numbers, duplicated logic
• Accessibility: Missing ARIA labels, keyboard navigation issues
• Edge Cases: Empty arrays, null inputs, boundary conditions

ℹ️ SUGGESTIONS (Nice to Have)

• Best Practices: Inconsistent naming, missing types, outdated patterns
• Optimization: Memoization opportunities, lazy loading, code splitting
• Maintainability: Complex logic without comments, long functions

🎨 Example Output

📊 Code Review Results

🚨 Critical: 2 | ⚠️  Warnings: 3 | ℹ️  Info: 1
────────────────────────────────────────────────────────

🚨 CRITICAL ISSUES (Must Fix)

1. 🚨 Potential runtime error: chained property access after JSON.parse without validation
   📁 File: src/api.ts (line 45)
   📝 Code: const userId = JSON.parse(response).user.id;
   ✅ Fix: Assign JSON.parse result to a variable, validate it, then access properties safely:
           const data = JSON.parse(response);
           if (data && typeof data === 'object') { /* use data */ }

2. 🔒 Security: Hardcoded secret detected
   📁 File: src/config.ts (line 12)
   📝 Code: const API_KEY = "sk-123456789";
   ✅ Fix: Move secrets to environment variables (.env) and add .env to .gitignore

────────────────────────────────────────────────────────

⚠️  WARNINGS (Recommended Fixes)

1. ⚠️ Potential memory leak: addEventListener without cleanup
   📁 File: src/components/Modal.tsx (line 28)
   📝 Code: window.addEventListener('keydown', handleEscape);
   💡 Fix: Store the listener reference and remove it in cleanup

────────────────────────────────────────────────────────
🛑 Commit blocked due to critical issues. Please fix them first.

⚙️ Configuration

Edit prefl.json in your project root:

{
  "ignore": {
    "globs": [
      "node_modules/**",
      "dist/**",
      ".git/**",
      "*.test.js",
      "coverage/**"
    ]
  },
  "review": {
    "blockSeverities": ["critical"],
    "context": {
      "baseLimit": 10,
      "importExpansionLimit": 20
    }
  }
}

Configuration Options

• ignore.globs: Files/folders to skip (supports glob patterns)
• review.blockSeverities: Which severity levels block commits (["critical"], ["critical", "warning"], or [])
• review.context.baseLimit: How many files to include as context (default: 10)
• review.context.importExpansionLimit: Max files to add via import tracking (default: 20)

🔧 Advanced Features

Generate and Apply Fixes

# Generate a patch file
prefl fix

# Validate the patch
prefl fix --dry-run

# Auto-apply the patch
prefl fix --apply

CI/CD Integration

# .github/workflows/code-review.yml
name: Prefl Code Review
on: [pull_request]

jobs:
  review:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - uses: actions/setup-node@v4
        with: { node-version: 20 }
      - run: npm i -g @preflight-ai/cli
      - run: prefl analyze --format json
        env:
          GROQ_API_KEY: ${{ secrets.GROQ_API_KEY }}

🌟 Why Prefl?

🤝 Contributing

We welcome contributions! Please see our Contributing Guide.

📄 License

Apache 2.0 - See LICENSE for details

💬 Support

• 🌐 Website: prefl.run
• 📧 Email: support@prefl.run
• 🐛 Issues: GitHub Issues

Made with ❤️ by developers, for developers