LangChain Agent MCP Server

# API Reference Complete API documentation for the LangChain Agent MCP Server. ## Base URL **Production:** https://langchain-agent-mcp-server-554655392699.us-central1.run.app **Local:** http://localhost:8000 ## Endpoints ### GET / Server information and available endpoints. **Response:** ```json { "name": "LangChain Agent MCP Server", "version": "1.0.0", "status": "running", "endpoints": { "manifest": "/mcp/manifest", "invoke": "/mcp/invoke" } } ``` ### GET /health Health check endpoint. **Response:** ```json { "status": "healthy" } ``` ### GET /mcp/manifest Returns the MCP manifest declaring available tools. **Response:** ```json { "name": "langchain-agent-mcp-server", "version": "1.0.0", "description": "LangChain Agent MCP Server...", "tools": [ { "name": "agent_executor", "description": "Execute a complex, multi-step reasoning task...", "inputSchema": { "type": "object", "properties": { "query": { "type": "string", "description": "The user's query or task" } }, "required": ["query"] } } ] } ``` ### POST /mcp/invoke Executes the LangChain agent with a user query. **Request:** ```json { "tool": "agent_executor", "arguments": { "query": "What is the capital of France?" } } ``` **Success Response (200):** ```json { "content": [ { "type": "text", "text": "The capital of France is Paris." } ], "isError": false } ``` **Error Response (400/500):** ```json { "content": [ { "type": "text", "text": "Error message here" } ], "isError": true } ``` **Error Codes:** - `400` - Bad Request (missing query, unknown tool) - `401` - Unauthorized (invalid API key) - `500` - Internal Server Error (agent execution failed) ### GET /docs Interactive API documentation (Swagger UI). Visit: https://langchain-agent-mcp-server-554655392699.us-central1.run.app/docs ## Authentication ### Optional API Key Authentication If `API_KEY` environment variable is set, include in requests: ```http Authorization: Bearer your-api-key ``` ## Rate Limiting Currently no rate limiting is implemented. Consider adding rate limiting for production use. ## Error Handling All errors follow the MCP response format: ```json { "content": [ { "type": "text", "text": "Error description" } ], "isError": true } ``` ## Examples ### cURL ```bash # Health check curl https://langchain-agent-mcp-server-554655392699.us-central1.run.app/health # Get manifest curl https://langchain-agent-mcp-server-554655392699.us-central1.run.app/mcp/manifest # Invoke agent curl -X POST https://langchain-agent-mcp-server-554655392699.us-central1.run.app/mcp/invoke \ -H "Content-Type: application/json" \ -d '{"tool":"agent_executor","arguments":{"query":"What is 2+2?"}}' ``` ### Python ```python import requests url = "https://langchain-agent-mcp-server-554655392699.us-central1.run.app/mcp/invoke" response = requests.post( url, json={ "tool": "agent_executor", "arguments": {"query": "What is 2+2?"} } ) print(response.json()) ``` ### JavaScript ```javascript fetch('https://langchain-agent-mcp-server-554655392699.us-central1.run.app/mcp/invoke', { method: 'POST', headers: { 'Content-Type': 'application/json' }, body: JSON.stringify({ tool: 'agent_executor', arguments: { query: 'What is 2+2?' } }) }) .then(res => res.json()) .then(data => console.log(data)); ``` --- For more examples, see [Examples](examples.md).