nasa

This NASA MCP Server exposes a suite of tools that give you access to NASA open APIs from a single MCP endpoint. It aggregates content across Astronomy Picture of the Day (APOD), Mars Rover Photos, Near Earth Objects (Asteroids), and the NASA Media Library, with caching and rate-limit awareness to improve reliability. You can query today’s APOD, fetch APODs over a date range, retrieve Mars rover images by sol or Earth date, browse asteroid data, and search NASA media assets. The server includes helpers for development and testing, as well as a collection of prompts and usage examples to help you build downstream applications and chat-based interfaces.

Prerequisites:

• Python 3.8+
• A NASA API key (free at api.nasa.gov)
• Python dependencies listed in requirements.txt

Install and run:

1. Clone the repository:

git clone https://github.com/jezweb/nasa-mcp-server.git
cd nasa-mcp-server

2. Install dependencies:

pip install -r requirements.txt

3. Configure environment variables:

cp .env.example .env
# Edit .env to include your NASA API key and desired settings

4. Run the server (STDIO transport by default):

python nasa_server.py

5. Optional: Run with HTTP transport (if configured):

TRANSPORT=http PORT=8000 python nasa_server.py

6. For development with FastMCP (if you use it):

fastmcp dev nasa_server.py

7. Test the server locally:

python nasa_server.py --test

Tips and notes:

• Ensure your NASA API key is configured in the .env file to avoid request failures.
• The server caches image data for 30 minutes and dynamic data for 10 minutes to balance freshness and latency.
• The TRANSPORT and LOG_LEVEL settings can be adjusted to fit your deployment environment (e.g., cloud, container, or local development).
• If you hit API rate limits, consider using multiple keys or raising the DEMO_HOURLY_LIMIT in your environment as appropriate for testing.
• The server supports multiple transport modes; default is STDIO, but HTTP transport can be enabled via environment variables.
• When deploying behind a reverse proxy, ensure the port mapping matches your deployment configuration.