MCP Census Server

What is MCP Census Server?

MCP Census Server is a Model Context Protocol (MCP) server that provides access to US Census Bureau data. It implements JSON-RPC 2.0 over HTTP, uses API key authentication, and offers a single tool (census/get) for retrieving census data. The server is designed for developers who want to integrate Census Bureau data into MCP-compatible applications or AI agents.

How to use MCP Census Server?

Clone the repository, install dependencies (pip install -r requirements.txt), set your Census API key and server API key in a .env file, then start the server with uvicorn app:app --host 0.0.0.0 --port 8000. Send JSON-RPC POST requests to http://localhost:8000/ with the X-API-Key header. The server supports the MCP methods initialize, notifications/initialized, tools/list, and tools/call.

Key features of MCP Census Server

• JSON-RPC 2.0 (single and batch requests)
• Pydantic models for request/response validation
• Caching of Census API responses (TTL 5 minutes)
• API key authentication via X-API-Key header
• Health check endpoint (/health)
• Interactive API docs (Swagger UI at /docs)

Use cases of MCP Census Server

• Programmatic retrieval of US Census Bureau data
• Integrate census data into MCP-compatible AI agents or chatbots
• Build client applications (e.g., the included Next.js MCP client) that query Census datasets
• Serve as a backend for data analysis workflows requiring recent Census statistics

FAQ from MCP Census Server

What Census data endpoints are available?

The server exposes the census/get tool. It retrieves data from the US Census Bureau API, but the README does not list specific Census endpoints or datasets.

What authentication is required?

Requests must include the X-API-Key header with your server API key (set in SERVER_API_KEY environment variable). Additionally, a valid Census API key must be set in the CENSUS_API_KEY environment variable.

What are the runtime requirements?

Python 3, the dependencies in requirements.txt, and a .env file with the required API keys. The server runs via uvicorn.

Does the server cache responses?

Yes, Census API responses are cached with a time-to-live (TTL) of 5 minutes.

Is there a client application included?

Yes, the repository contains a Next.js-based MCP client in the client/ directory that can interact with this server.