Write Aid MCP - AI Checker

An intelligent text analysis tool with MCP (Model Context Protocol) integration for detecting AI-generated content.

🚀 Features

• Frontend UI: Clean, modern interface for text analysis
• Backend API: Flask-based server with multiple endpoints
• MCP Integration: Support for Model Context Protocol clients
• Finchat API: Optional integration with Finchat AI services
• Local Analysis: Heuristic-based fallback when API is unavailable
• Real-time Analysis: Instant feedback on text authenticity

📋 Prerequisites

• Python 3.8 or higher
• Modern web browser (Chrome, Firefox, Safari, Edge)
• Git (for development)

🔧 Installation

1. Clone the repository:

   git clone https://github.com/MikeVenge/write-aid-mcp.git
   cd write-aid-mcp

2. Install dependencies:

   pip3 install -r requirements.txt

3. Configure (Optional):

   • Edit ai_checker_config.js for frontend settings
   • Set environment variables for Finchat API (see below)

🎯 Quick Start

Option 1: Run with Backend (Recommended)

./start_with_backend.sh

This will:

• Start the backend server on port 5001
• Start the frontend server on port 8000
• Open the app in your browser

Option 2: Run Backend and Frontend Separately

Terminal 1 - Backend:

python3 backend_server.py

Terminal 2 - Frontend:

python3 -m http.server 8000

Browser:

http://localhost:8000/ai_checker.html

🔐 Finchat API Configuration (Optional)

To use the Finchat API integration, set these environment variables:

export FINCHAT_BASE_URL="https://your-finchat-instance.com"
export FINCHAT_API_TOKEN="your_jwt_token_here"
export FINCHAT_COT_SLUG="ai-detector"
export FINCHAT_MODEL="gemini-2.5-flash"

See QUICK_SETUP.md for detailed configuration.

📁 Project Structure

write-aid-mcp/
├── ai_checker.html          # Main frontend interface
├── ai_checker.js            # Frontend JavaScript logic
├── ai_checker.css           # Styling
├── ai_checker_config.js     # Frontend configuration
├── backend_server.py        # Flask backend server
├── mcp_*.py                 # MCP client implementations
├── requirements.txt         # Python dependencies
├── start_with_backend.sh    # Startup script
└── docs/                    # Documentation files

🧪 Testing

Run the test suite:

python3 test_ai_detector.py

📚 Documentation

• RUN_INSTRUCTIONS.md - Detailed running instructions
• BACKEND_SETUP.md - Backend configuration guide
• QUICK_SETUP.md - Quick reference guide
• TROUBLESHOOTING.md - Common issues and solutions
• MCP_INTEGRATION_GUIDE.md - MCP protocol details

🔍 How It Works

1. User Input: Paste text into the web interface
2. Analysis: Text is analyzed by:
   • Finchat API (if configured)
   • Local heuristic algorithms (fallback)
3. Results: Display confidence score and detailed analysis
4. Feedback: Visual indicators for AI vs. human-written content

🛠️ API Endpoints

Backend Server (Port 5001)

• GET /health - Server health check
• POST /analyze - Analyze text for AI content
• POST /mcp/analyze - MCP-compatible analysis endpoint

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

📄 License

MIT License - see LICENSE file for details

👤 Author

Mike Venge

• GitHub: @MikeVenge
• Repository: write-aid-mcp

🐛 Issues

Found a bug? Please open an issue on GitHub Issues

⭐ Support

If you find this project helpful, please give it a star on GitHub!

---

Version: 1.0.0
Last Updated: November 10, 2025