MeteoSwiss MCP Server

A Model Context Protocol (MCP) server that exposes Swiss weather forecast data callable tools.
It fetches data from the official MeteoSwiss meteodata-lab, caches it locally, and serves predictions such as rainfall, sunshine, temperature, etc. The prediction data is from the ICON-CH2-EPS forecast system that produces data for up to 5 days ahead.

Additionally there is also a MCP client that can be run to test the server using the stdio transport.

Note:
This project is not an official MeteoSwiss product.
All forecast data are from the MeteoSwiss Open Data portal.
Source: MeteoSwiss

📦 Installation

# Clone the repo
git clone https://github.com/cuolm/meteo-swiss-mcp.git
cd meteo-swiss-mcp

# Create a virtual environment 
python -m venv .venv
source .venv/bin/activate   # Windows: .\.venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

Note:

• Ollama is optional – only needed if you want to use the MCP client meteo_swiss_mcp_client.py.

• The server uses a cache (cache/EarthKitCache) to avoid re‑downloading weather data. Clear it with rm -rf cache/EarthKitCache if needed.

• The server uses a cache (cache/nominatim_geocode_cache.json) for lat/lon lookups. Clear it with rm -rf cache/nominatim_geocode_cache.json if needed.

⚙️ Server Configuration

Create a .env file in the root directory and specify an environment variable that tells Nominatim (the geocoding service) who is making the call.

echo 'NOMINATIM_USER_AGENT="YourWeatherMCPServer/1.0 (yourname@example.com)"' > .env

🚀 Running the Server

# stdio (default)
python src/meteo_swiss_mcp_server.py

# streamable-http
python src/meteo_swiss_mcp_server.py --transport=streamable-http --host=localhost --port=8050

Optional flags: --help

🐳 Running the Server with Docker

Run the MCP server in Docker with these steps:

1. Create a .env file in the project root containing your Nominatim user agent environment variable (replace "YourWeatherMCPServer/1.0 (yourname@example.com)"):

   echo 'NOMINATIM_USER_AGENT="YourWeatherMCPServer/1.0 (yourname@example.com)"' > .env

2. Build the Docker image from the root folder:

   docker build -t meteo_swiss_mcp_server .

3. Run the container, passing the .env file and mapping port 8050:

   docker run --env-file .env -p 8050:8050 meteo_swiss_mcp_server

4. Access the server at:

   http://localhost:8050/mcp/

This runs the MCP server isolated with all dependencies and environment variables preconfigured.

🖥️ Running the MCP Client using Stdio Transport

The MCP client src/meteo_swiss_mcp_client.py can be used to test the server over the stdio transport. Make sure Ollama is installed on your system. You can download it here or install via Homebrew on macOS: brew install ollama

# Pull a local Ollama LLM model (e.g. qwen3:4b)
ollama pull qwen3:4b

# Run the MCP client (the client script will automatically start the server)
python src/meteo_swiss_mcp_client.py --model=qwen3:4b --server-script=src/meteo_swiss_mcp_server.py

🔧 Available Tools

• Lead time is the number of hours counted from Swiss local time 00:00, internally converted to UTC (the ICON-CH2-EPS forecast system uses UTC).

• Example: A lead time of 36 hours returns the forecast for 12:00 Swiss local time tomorrow.

• Minimum lead time: 2 hours; maximum lead time: 121 hours.

📂 Project Structure

meteo-swiss-mcp/
├── src/
│   ├── meteo_swiss_mcp_server.py   # MCP server
│   ├── meteo_swiss_predictions.py  # Data fetching logic
│   └── meteo_swiss_mcp_client.py   # MCP client (optional)
├── requirements.txt
├── .env
│
├── cache/                           
├── tests/                           
├── docs/                           
└── Dockerfile                           

⚡ Example Usage with LMStudio

Using the streamable-http transport layer

Configure the mcp.json file in LMStudio:

{
  "mcpServers": {
    "meteo_swiss_mcp_server": {
      "url": "http://localhost:8050/mcp/"
    }
  }
}

Run the MCP server with the streamable-http transport layer:

# Make sure the virtual environment is activated 
source .venv/bin/activate

python src/meteo_swiss_mcp_server.py --transport=streamable-http --host=localhost --port=8050

Using the stdio transport layer

Configure the mcp.json file in LMStudio. Replace <path-to-the-project> with your actual local path:

{
  "mcpServers": {
    "meteo_swiss_mcp_server": {
      "command": "<path-to-the-project>/.venv/bin/python",
      "args": [
        "<path-to-the-project>/src/meteo_swiss_mcp_server.py"
      ]
    }
  }
}

🧪 Tests

Run all tests with:

cd tests
pytest

📚 Useful Resources

• Meteo Swiss Open Data

• Jupyter Notebook Examples

• Model Context Protocol

• MCP Server Quickstart

📄 License

This project is licensed under the Apache License 2.0.