Skip to content
Start free
Start free
Explore

Try CrateDB Live: Connect

MCP Server
  • 1. Choose Scenario
  • 2. Get Ready
  • 3. Run CrateDB
  • 4. Import Data
  • 5. Explore Queries
  • 6. More Queries
  • 7. Connect
  • 8. Next Steps

This guide builds a small MCP server over the dataset. In a few minutes you will have an assistant that answers questions such as "What was the coldest place in Germany yesterday?" against live data.

The full scenario is explained in more details here.

Install and connect the MCP server

  • Copyi the german_weather_mcp.py file to your project.
  • You may need to run "pip install --quiet "mcp[cli]" httpx", depending on your environment.
  • Update the .mcp.json file at the root of your project.
 
{
  "mcpServers": {
    "german-weather": {
      "command": "python",
      "args": [
        "/path/to/german_weather_mcp.py",
        "--cratedb-url",
        "http://:@:4200/"
      ]
    }
  }
}
  • Restart Claude and check that the MCP server is connect (using the /mcp command)
  • Ask Claude: "please check my german-weather" is properly configured in .mcp.json

Use the MCP server

  • Ask Claude: "how to call my german-weather mcp server?"
  • Ask Claude: "which questions can I ask the german weather mcp?"
  • Ask for example "What's the coldest temperature ever recorded in Germany?"