AI Readiness Audits - A comprehensive toolkit for auditing websites for AI readiness, SEO optimization, and content quality.

This monorepo contains:

• packages/scanner - Core scanning engine with 50+ rules for AI readiness detection
• packages/utils - Shared utilities across packages

• apps/cli - Command-line interface with interactive wizard and preset configurations
• apps/web - Next.js web application for visual audits
• apps/api - Express API server for programmatic access

pnpm install

# Interactive wizard (simplest way)
pnpm --filter=@ai-lighthouse/cli dev audit https://example.com

# Using presets (recommended)
pnpm --filter=@ai-lighthouse/cli dev audit https://example.com --preset ai-optimized

# List available presets
pnpm --filter=@ai-lighthouse/cli dev presets

# Start the web app
pnpm --filter=@ai-lighthouse/web dev

# Visit http://localhost:3000

# Start the API server
pnpm --filter=@ai-lighthouse/api dev

# API available at http://localhost:3001

• 50+ Built-in Rules covering:

  • AI Readiness & Comprehension
  • Content Clarity & Quality
  • Extractability & Chunking
  • Knowledge Graph & Schema.org
  • SEO & Crawlability
  • Accessibility (A11y)
  • Security & Performance

• LLM Integration:

  • Support for OpenAI, Anthropic, Ollama, and local models
  • Hallucination detection
  • Content comprehension analysis
  • Entity extraction
  • FAQ generation

• Advanced Analysis:

  • Content chunking with token optimization
  • Extractability mapping
  • AI readiness scoring
  • Traditional SEO scoring

Interactive Features:

• Beautiful terminal UI built with React (Ink)
• Interactive wizard for easy configuration
• Preset configurations (basic, ai-optimized, full, minimal)
• Tab navigation through analysis sections

Commands:

1. audit - Audit a single webpage with presets
2. crawl - Crawl and audit multiple pages
3. report - Generate and view reports
4. presets - List available preset configurations

Output Formats:

• Interactive terminal UI (default)
• JSON (CI/CD friendly)
• HTML (beautiful visualizations)
• PDF (for sharing)
• CSV (spreadsheet analysis)

• Visual audit interface built with Next.js
• Real-time AI readiness analysis
• Interactive charts and visualizations
• Tab-based navigation through results
• Export capabilities

• RESTful API for programmatic access
• Rate limiting and security features
• Webhook support for automation
• OpenAPI documentation

• packages/scanner/README.md - Scanner API documentation
• packages/scanner/SCORING.md - Scoring system explained
• apps/cli/README.md - CLI documentation and commands
• apps/cli/EXAMPLES.md - Usage examples and recipes
• apps/web/README.md - Web application documentation
• apps/api/README.md - API server documentation

This project uses Turborepo for managing the monorepo.

# Install dependencies
pnpm install

# Build all packages
pnpm build

# Run specific package in dev mode
pnpm --filter=@ai-lighthouse/cli dev
pnpm --filter=@ai-lighthouse/web dev
pnpm --filter=@ai-lighthouse/api dev

# Run tests
pnpm test

# Lint all packages
pnpm lint

ai-lighthouse/
├── packages/
│   ├── scanner/          # Core scanning engine
│   │   ├── src/
│   │   │   ├── rules/    # 50+ detection rules
│   │   │   ├── llm/      # LLM integration
│   │   │   └── scoring/  # AI readiness scoring
│   └── utils/            # Shared utilities
└── apps/
    ├── cli/              # Command-line interface
    │   ├── src/
    │   │   ├── commands/ # CLI commands
    │   │   ├── ui/       # Interactive UI components
    │   │   └── presets.ts # Preset configurations
    ├── web/              # Next.js web app
    └── api/              # Express API server

╔════════════════════════════════════════════════════════════════╗
║           AI Readiness Assessment for example.com              ║
╚════════════════════════════════════════════════════════════════╝

📊 Overall AI Readiness: 87/100 (B+)

🔍 Extractability    : 92/100 ⭐⭐⭐⭐⭐
📝 Clarity          : 85/100 ⭐⭐⭐⭐
🧩 Chunkability     : 88/100 ⭐⭐⭐⭐
🎯 Context          : 84/100 ⭐⭐⭐⭐

Beautiful, interactive reports with:

• Color-coded severity levels
• Score visualizations
• Detailed issue breakdowns
• Entity detection display
• Actionable remediation steps

# GitHub Actions example
- name: Audit Website
  run: |
    pnpm --filter=@ai-lighthouse/cli dev audit ${{ secrets.SITE_URL }} \
      --preset minimal \
      --threshold 80 \
      --output json

Exit code 1 if score is below threshold - perfect for automated quality gates!

MIT

Contributions welcome! Please see individual package READMEs for specific contribution guidelines.