Meteoswiss

You can run a dedicated MCP server that exposes MeteoSwiss weather data to AI assistants. This server provides multilingual weather reports, regional forecasts, and search/fetch capabilities, all accessible through a standard MCP endpoint for real-time interactions and remote integrations.

How to use

Connect to the MCP endpoint to start a session and receive weather data through your MCP client. The server serves an HTTP endpoint for the MCP stream and a separate health endpoint for monitoring, so you can integrate it into your assistant workflow and verify its status at any time. When you want to run the server locally, use the recommended start command and point your MCP client to the SSE endpoint at /mcp.

How to install

Prerequisites: you need Node.js v18.0.0 or later and a package manager such as pnpm. If you are unsure about your Node version, you can manage it with a tool like nvm.

Step-by-step setup to run the server locally:

Additional configuration and usage notes

The server runs as an HTTP service with Server-Sent Events (SSE) for real-time communication. It exposes these endpoints: GET / for server information, GET /mcp for MCP SSE connections, POST /messages?sessionId=... for message handling, and GET /health for a health check.

Example run flow

# Step 1: clone the project
git clone https://github.com/eins78/mcp-server-meteoswiss-data.git
cd mcp-server-meteoswiss-data

# Step 2: install dependencies
pnpm install

# Step 3: start the server (default port 3000)
pnpm start

# Optional: run on a different port
PORT=8080 pnpm start

Available tools

meteoswissWeatherReport

Get official MeteoSwiss weather reports for Swiss regions with multilingual support and daily forecasts.

search

Search MeteoSwiss website content with pagination and multi-language support.

fetch

Retrieve full content from MeteoSwiss pages in various formats (markdown, text, HTML).