Skip to main content
Glama
NO_API_KEYS.md2.23 kB
# No API Keys Required! ## How It Works The MCP Router **does NOT need API keys** - it only recommends which model Cursor should use. Cursor handles all API calls with its own configured keys. ``` Your Query → Router Analyzes → Recommends Model → Cursor Uses Model (with its own API keys) ``` ## Benefits ✅ **No API key management** - Cursor already has your keys configured ✅ **Simpler setup** - Just install MCP SDK and configure Cursor ✅ **Secure** - API keys stay in Cursor, not in the router ✅ **Flexible** - Router can recommend models without needing access to them ## Setup (No API Keys Needed) 1. **Install MCP SDK:** ```bash pip3 install --user mcp ``` 2. **Run setup script:** ```bash ./scripts/setup_cursor.sh ``` (No API keys needed in the config!) 3. **Restart Cursor** - That's it! ## How Cursor Uses It When you type a query in Cursor: 1. **Router analyzes** your query (task type, complexity, etc.) 2. **Router recommends** the best model (e.g., "Use GPT-4o-mini for this code task") 3. **Cursor receives** the recommendation 4. **Cursor makes API call** using the recommended model with its own API keys 5. **You get response** from the optimally selected model ## MCP Tools Available - `route_query` - Get routing decision with reasoning - `get_model_recommendation` - Get model ID for Cursor to use - `list_models` - See all available models - `analyze_query` - Analyze query characteristics - `get_routing_stats` - View routing statistics ## Example Flow **You:** "Write a Python function to sort a list" **Router:** ```json { "recommended_model": { "model_id": "gpt-4o-mini", "provider": "openai", "name": "GPT-4o-mini" }, "confidence": 0.92, "reasoning": "Model is optimized for code_generation tasks; Selected for low latency" } ``` **Cursor:** Uses `gpt-4o-mini` with its own OpenAI API key → Returns response ## Configuration The MCP config doesn't need API keys: ```json { "version": "1.0", "mcpServers": { "mcp-router": { "command": "python3", "args": ["/path/to/mcp-router/src/mcp_server.py"], "env": {} } } } ``` That's it! No API keys needed because Cursor handles all the API calls.

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/AI-Castle-Labs/mcp-router'

If you have feedback or need assistance with the MCP directory API, please join our Discord server