Context7-ChatGPT-Bridge

# Context7 ChatGPT Bridge ## Why This Exists ChatGPT needs access to current programming documentation but requires specific `search` and `fetch` tools. Context7 provides excellent up-to-date docs but uses different tools (`resolve-library-id` and `get-library-docs`). This bridge translates between them, giving ChatGPT access to Context7's documentation database. ## What It Does A bridge that allows ChatGPT to access up-to-date programming documentation through the Context7 MCP server. Implements ChatGPT's required `search` and `fetch` tools while using Context7's documentation database internally. ## Requirements - **Node.js** >= 18.0.0 - **Python** >= 3.8 - **ngrok** (for ChatGPT access) ## Quick Start 1. **Install dependencies:** ```bash pip install -r requirements.txt ``` 2. **Start the bridge:** ```bash python context7_bridge.py ``` The script automatically starts ngrok and displays the ChatGPT-ready URL. 3. **Add to ChatGPT:** Copy the displayed URL to ChatGPT's MCP connectors: ``` https://abc123.ngrok-free.app/sse ``` ## Available Tools ### `search` Search for programming libraries and frameworks. Supports both library names (e.g., "React", "MongoDB") and direct Context7 library IDs (e.g., "/reactjs/react.dev"). ### `fetch` Fetch comprehensive documentation for specific libraries. Supports advanced parameters: - **Basic:** `library_id` - **Topic-focused:** `library_id|topic:hooks` - **Custom tokens:** `library_id|tokens:15000` - **Combined:** `library_id|topic:authentication|tokens:12000` ### Example Topics - `hooks` - React hooks, useEffect, useState - `routing` - Navigation, route setup - `authentication` - Login, security, JWT - `installation` - Setup, configuration - `api` - API reference, methods - `examples` - Code examples, tutorials ## How It Works ``` ChatGPT → ngrok → Bridge → Context7 MCP Server → Documentation Database ``` 1. ChatGPT sends search/fetch requests to your bridge 2. Bridge translates these to Context7's resolve-library-id/get-library-docs calls 3. Context7 returns current documentation 4. Bridge formats responses for ChatGPT ## Configuration **Command-line options:** ```bash python context7_bridge.py --help ``` - `--port` - Port to run on (default: 8000) - `--host` - Host to bind to (default: 127.0.0.1) - `--no-ngrok` - Disable automatic ngrok tunnel **Environment variables:** - `LOG_LEVEL` - Logging level (default: INFO) ## Manual Setup If you prefer manual ngrok control: ```bash # Start without ngrok python context7_bridge.py --no-ngrok # In another terminal ngrok http 8000 ``` ## Testing Test without ChatGPT: ```bash # Health check curl http://localhost:8000/health # Test search curl -X POST http://localhost:8000/sse \ -H "Content-Type: application/json" \ -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"search","arguments":{"query":"react"}}}' ``` ## Troubleshooting **"Could not get response from Context7 server"** - Ensure Node.js and npx are installed and in PATH **"Unknown document ID"** - Always call `search` before `fetch` to get valid IDs - Or use direct Context7 library IDs (starting with `/`) - Note: Tool descriptions may need refinement for ChatGPT to better understand the search-first workflow - currently works but ChatGPT occasionally has hiccups with the sequence **Debug mode:** ```bash LOG_LEVEL=DEBUG python context7_bridge.py ```