Prefl AI - Professional Code Review Tool ✨

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

Website: prefl.run

🚀 Features

• 🤖 AI-Powered Analysis - Deep semantic code review with Groq LLM (senior architect-level analysis)
• 🐛 Production Bug Detection - Catches critical bugs that pass code review & testing but crash in production
• 🎯 Smart Context Tracking - Analyzes related code and imports to maintain full context
• 🚨 Critical Issue Detection - Identifies production-critical problems
• 🌍 Multi-Language Support - Works with any programming language (JavaScript, Python, Go, Java, C#, Ruby, PHP, Rust, etc.)
• ⚡ Lightning Fast - AI analysis in 2-5 seconds
• 🪝 Automatic Pre-commit Hook - Installs git pre-commit hook automatically on init

📦 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
• ✅ Automatically install pre-commit hook (if git repository)

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 (Pre-commit Hook)

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 (default)
prefl analyze

# Analyze entire project
prefl analyze --full

# Save results to prefl-log.txt
prefl analyze --output

Fix Issues

# Generate AI-suggested patch (saves to prefl-log.txt automatically)
prefl fix

# Apply the patch automatically
prefl fix --apply

# Validate patch without applying
prefl fix --dry-run

📊 How It Works

Prefl uses Groq AI to analyze your code like a professional code reviewer and tester:

1. Analyzes staged files - Reviews the code you're changing
2. Checks related context - Examines imports, dependencies, and connected files
3. Tests mentally - Thinks through various scenarios and edge cases
4. Finds issues - Identifies bugs, critical problems, and potential issues
5. Reports findings - Provides detailed explanations and fix suggestions

Context-Aware Analysis

Prefl automatically:

• Loads staged files (the code being changed)
• Finds related files via imports and dependencies
• Maintains full context to understand how code interacts with the rest of the system
• Analyzes compatibility to ensure changes don't break related code

📋 Commands

prefl init                # Setup Prefl in your project (installs pre-commit hook automatically)
prefl analyze             # Analyze staged changes
prefl analyze --full      # Analyze entire project
prefl analyze --output    # Analyze and save results to prefl-log.txt
prefl fix                 # Generate AI-suggested patches (saves to prefl-log.txt)
prefl fix --apply         # Auto-apply generated fixes
prefl fix --dry-run       # Validate patch without applying
prefl --version           # Show version
prefl --help              # Show help

🎨 Example Output

🤖 Analyzing with AI (5 staged files + 12 related files for context)...
✨ Analysis complete! Found 3 issue(s)

📊 Code Review Results

🚨 Critical: 3 | ⚠️  Warnings: 0 | ℹ️  Info: 0
────────────────────────────────────────────────────────

🚨 CRITICAL ISSUES (Must Fix)

1. ⚡ Race Condition
   File:     src/payment.ts:45
   Issue:    Race condition: shared state modified in async function without locking.
             Multiple concurrent requests will corrupt balance data.
   Fix:      Use db.transaction() or mutex.lock() to ensure atomicity

2. 🔒 Security
   File:     src/api/user.ts:23
   Issue:    Missing authentication middleware on /admin/users endpoint.
             Anyone can access admin-only data.
   Fix:      Add requireAuth and requireAdmin middleware before handler

3. 🧮 Business Logic
   File:     src/utils.ts:89
   Issue:    Function returns undefined for non-premium users.
             Missing else branch causes caller code to break.
   Fix:      Add explicit return for all code paths

⚙️ Configuration

Edit prefl.json in your project root:

{
  "ignore": {
    "globs": [
      "node_modules/**",
      "dist/**",
      ".git/**",
      "*.test.js",
      "coverage/**"
    ]
  },
  "review": {
    "blockSeverities": ["critical"],
    "context": {
      "importExpansionLimit": 50
    }
  }
}

Configuration Options

• ignore.globs: Files/folders to skip (supports glob patterns)
• review.blockSeverities: Which severity levels block commits (["critical"], ["critical", "warning"], or [])
• review.context.importExpansionLimit: Max files to add via import tracking (default: 50)

🌟 Why Prefl?

Prefl uses AI to catch bugs that humans miss - issues that look correct in review, pass tests, but crash in production.

The AI analyzes:

• Context - How code interacts with related files and dependencies
• Edge cases - Boundary conditions, null checks, error handling
• Security - Authentication, authorization, input validation
• Business logic - Completeness, missing returns, undefined results
• And more - Anything that could cause problems in production

🤝 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