Coinbase Chat MCP

# Coinbase Chat MCP - Deployment Guide ## 🚀 Quick Deploy to Vercel ### Prerequisites - [Vercel account](https://vercel.com) - [OpenAI API key](https://platform.openai.com/api-keys) (for AI features) ### 1. Deploy to Vercel [](https://vercel.com/new/clone?repository-url=https://github.com/your-username/coinbase-chat-mcp) Or manually: ```bash # Install Vercel CLI npm i -g vercel # Deploy from project root cd frontend vercel --prod ``` ### 2. Configure Environment Variables In your Vercel dashboard, add these environment variables: | Variable | Value | Required | |----------|-------|----------| | `VITE_OPENAI_API_KEY` | `sk-your-openai-api-key-here` | Yes (for AI features) | | `VITE_API_URL` | `https://your-backend-url.vercel.app` | Optional | ### 3. Backend Deployment The MCP server and API backend can be deployed separately: ```bash # Deploy API server cd api-server vercel --prod # Deploy MCP server (if needed separately) cd mcp-server vercel --prod ``` ## 🔒 Security Features Implemented ### Rate Limiting - **Limit**: 3 requests per minute per user - **Storage**: Client-side localStorage (educational purposes) - **Reset**: Automatic after 1 minute window - **UI**: Visual indicator showing remaining requests ### Conversation Guardrails - **Allowed Topics**: Cryptocurrency, blockchain, trading, MCP technology - **Blocked Topics**: Pets, weather, food, entertainment, etc. - **Behavior**: Politely redirects off-topic conversations - **Implementation**: Keyword-based filtering with smart detection ## 🎓 Educational Configuration ### Rate Limiting Configuration ```typescript // In aiService.ts const RATE_LIMIT_MAX_REQUESTS = 3; const RATE_LIMIT_WINDOW_MS = 60 * 1000; // 1 minute ``` ### Topic Guardrails ```typescript // Allowed topics const ALLOWED_TOPICS = [ 'cryptocurrency', 'crypto', 'bitcoin', 'ethereum', 'blockchain', 'trading', 'price', 'market', 'analysis', 'mcp', 'coinbase' ]; // Blocked topics const OFF_TOPIC_KEYWORDS = [ 'cat', 'dog', 'animal', 'weather', 'food', 'movie', 'music' ]; ``` ## 🛠️ Customization ### Adjust Rate Limits To modify rate limits for your deployment: 1. Edit `frontend/src/services/aiService.ts` 2. Change `RATE_LIMIT_MAX_REQUESTS` and `RATE_LIMIT_WINDOW_MS` 3. Redeploy to Vercel ### Modify Topic Guardrails To adjust conversation topics: 1. Edit `ALLOWED_TOPICS` and `OFF_TOPIC_KEYWORDS` arrays 2. Modify the `isTopicAllowed()` method logic 3. Update system prompts in the AI service ### Environment Variables for Customization ```bash # Optional: Custom rate limiting (if implemented) VITE_RATE_LIMIT_MAX_REQUESTS=5 VITE_RATE_LIMIT_WINDOW_MS=120000 # Optional: Custom API endpoints VITE_API_URL=https://your-custom-api.com ``` ## 📊 Monitoring & Analytics ### Rate Limit Monitoring - Visual indicators in the UI - Console logging for debugging - localStorage tracking for persistence ### Usage Patterns - Topic filtering effectiveness - Rate limit hit frequency - User engagement metrics ## 🔧 Troubleshooting ### Common Issues 1. **Rate Limit Not Working** - Check localStorage is enabled - Verify browser compatibility - Clear localStorage if needed 2. **Guardrails Too Restrictive** - Add more allowed topics - Adjust keyword detection logic - Update system prompts 3. **OpenAI API Issues** - Verify API key is correct - Check API quota and billing - Ensure CORS is configured ### Debug Mode Enable debug logging: ```typescript // Add to aiService.ts const DEBUG = true; console.log('Rate limit check:', rateCheck); console.log('Topic allowed:', isAllowed); ``` ## 🚀 Production Recommendations 1. **Server-Side Rate Limiting**: Move to backend for production 2. **Database Storage**: Replace localStorage with Redis/DB 3. **Advanced Guardrails**: Implement ML-based content filtering 4. **Monitoring**: Add proper analytics and logging 5. **Caching**: Implement response caching for better performance ## 📝 License & Usage This project is designed for educational purposes. The rate limiting and guardrails are implemented to: - Demonstrate responsible AI usage - Protect API quotas during learning - Focus conversations on relevant topics - Provide a professional demo experience For production use, consider implementing proper server-side controls and authentication.