Swiss Weather

Swiss Weather MCP Server

A Model Context Protocol (MCP) server that provides AI applications with weather information for any place in Switzerland.

• Forecasts for today and the next 8 days from the MeteoSwiss local forecasts, updated every hour
• Measurements from about 160 SwissMetNet weather stations, updated every 10 minutes
• Places on the Swiss national map, such as towns, districts, lakes and peaks, from swissNAMES3D
• When the Sun and the Moon rise and set at a place, the civil twilight, the Moon's phase and how much of it is lit, calculated for any day
• A bundled MCP client for testing

Data sources. This project is not an official MeteoSwiss product.

• Weather forecasts and measurements: MeteoSwiss Open Data. Source: MeteoSwiss
• Place names: swissNAMES3D. Source: Federal Office of Topography swisstopo
• Municipalities and cantons: swissBOUNDARIES3D. Source: Federal Office of Topography swisstopo

Table of Contents

• Project Structure
• Installation
• Usage
• Available Tools
• Good to Know
• Example Usage with LMStudio
• Resources
• License

Project Structure

swiss-weather-mcp/
├── src/swiss_weather_mcp/
│   ├── server.py           # MCP server and its tools
│   ├── places/             # swisstopo place lookup
│   ├── forecast/           # MeteoSwiss local forecasts
│   ├── measurements/       # MeteoSwiss SwissMetNet measurements
│   ├── astronomy/          # Sunrise, sunset, moonrise, moonset and phase
│   ├── opendata.py         # Downloads from data.geo.admin.ch
│   ├── formatting.py       # Times, wind directions and height differences in the answers
│   ├── errors.py           # The error for questions the data cannot answer
│   ├── log.py              # Logging setup
│   └── client.py           # MCP client
├── tests/swiss_weather_mcp/  # Tests
└── .github/workflows/      # CI and release pipelines

Installation

# As a command line tool
uv tool install swiss-weather-mcp

# With the optional MCP client, which pulls in the OpenAI SDK
uv tool install 'swiss-weather-mcp[client]'

# As a library dependency
uv add swiss-weather-mcp  # or: pip install swiss-weather-mcp

From source:

git clone https://github.com/cuolm/swiss-weather-mcp.git
cd swiss-weather-mcp
uv sync --extra client  # or, in a virtual environment: pip install -e ".[client]"

No configuration is needed, the server reads everything it requires from MeteoSwiss and swisstopo.

Usage

Running the Server

# stdio (default)
swiss-weather-mcp-server

# streamable-http
swiss-weather-mcp-server --transport=streamable-http --host=localhost --port=8050

From a source checkout, put uv run in front. --help lists all flags.

Where the server keeps its files, and --cache-all-forecast-points: see Cache and Downloads.

Running the Server with Docker

Images are built and published automatically by GitHub Actions to the project's GitHub Container Registry, tagged :latest (newest release) and by version.

# Published image
docker run -p 8050:8050 ghcr.io/cuolm/swiss-weather-mcp:latest

# Built from a source checkout
docker build -t swiss-weather-mcp .
docker run -p 8050:8050 swiss-weather-mcp

The server is then at http://localhost:8050/mcp/.

Each docker run starts a new container with an empty cache. Mount a volume to keep it, for example -v swiss-weather-cache:/root/.cache/swiss-weather-mcp.

Running the MCP Client

The bundled MCP client tests the server over stdio. It needs the client extra (see Installation) and starts the MCP server itself, but not the LLM, which has to be running first, for example in llama-server.

# Install llama.cpp, which provides llama-server, here with Homebrew on macOS or Linux
brew install llama.cpp

# Start llama-server with a model in its own terminal; it downloads the model on first use
llama-server --jinja --no-mmproj -hf unsloth/Qwen3.5-4B-GGUF --port 8080

# Run the client against it in a second terminal
swiss-weather-mcp-client --model=unsloth/Qwen3.5-4B-GGUF --base-url=http://localhost:8080/v1

From a source checkout, put uv run --extra client in front of the client. Other ways to install llama.cpp are in its installation guide.

--jinja uses the model's chat template, which tool calling needs (the default in current llama.cpp), and --no-mmproj skips the vision projector. --base-url is the server's OpenAI compatible API, and --model the name it lists under /v1/models.

Available Tools

Times are Swiss local time (see Times). Every tool with a location also takes an optional near, needed only when several places share the name (see Locations). Answers name the forecast point when it is not next to the place (see Forecast Points).

Tool	Answer
forecast_daily(location, start_date, days)	1 to 9 days: lowest and highest temperature, rain, weather with an emoji
forecast_rain(location, start, end)	Chance and amount of rain in 3-hour blocks
forecast_temperature(location, start, end)	Temperature hour by hour, or for one hour
forecast_sunshine(location, start, end)	Hours of sunshine over a period, or in one hour
forecast_wind(location, when)	Mean speed, gusts and direction
forecast_cloud_cover(location, when)	Total cloud cover and its low, medium and high layers
forecast_freezing_level(location, when)	Height of the 0 °C line
measurements_current(location)	What a station near the place measures now (see Measurements)
measurements_extremes(extreme)	Where it is warmest, coldest, windiest, wettest or sunniest now
astronomy_sun(location, date)	Sunrise, sunset and civil twilight
astronomy_moon(location, date)	Moonrise, moonset, phase and how much is lit
current_date_and_time()	Today's weekday and the Swiss time now

Good to Know

Times

• Times are Swiss clock time in ISO 8601, such as "2026-09-23T14:00". The server applies summer or winter time.
• Answers give times with their UTC offset, such as 2026-09-23T13:00+02:00.
• An hourly value stamped 14:00 covers 13:00 to 14:00. Cloud cover and the freezing level are values at that moment.
• The forecast covers today and the next 8 days. current_date_and_time() gives today's date, which questions such as "tomorrow" or "tonight" need.

Locations

• A location is a place name from swissNAMES3D: a town, village, city district, peak, pass, lake, glacier, valley or railway station, such as Zürich, Oerlikon, Titlis or Zürichsee.
• Case and accents don't matter, and a few English names such as Geneva work too.
• A name of several places, such as Blausee, needs near: a village next to the one the user means, such as near="Riederalp". Without it, the tool lists all places of the name, the biggest first, each with the call that chooses it and the municipality and canton it lies in:

  'Blausee' is the name of 6 places. [...] The biggest first:
    location="Blausee", near="Blausee-Mitholz": district and lake, fewer than 20 people, 887 m, municipality Kandergrund, canton Bern
    location="Blausee", near="Ramiswil": settlement, fewer than 20 people, 683 m, municipality Mümliswil-Ramiswil, canton Solothurn
    ...
  

• The municipalities and cantons carry swisstopo's names, in the local language, such as Genève or Graubünden. They appear once the swissBOUNDARIES3D GeoPackage is downloaded in the background at start.
• An unknown name gets up to 10 suggestions, such as Davos Dorf and Davos Platz for Davos.

Forecast Points

• The forecast comes from a MeteoSwiss forecast point near the place. An answer names it in data_from when it is more than 1 km away or more than 50 m higher or lower.
• Temperature changes faster with height than with distance: in the standard atmosphere (ISO 2533), air cools by 0.65 °C per 100 m.
• So among the points within 5 km, the server takes the one with the smallest distance plus 20 times the height gap: 100 m of height gap counts like 2 km of distance. Without a point within 5 km, it takes the nearest one.
• The factor 20 was tested on the real data: up to 20, the height gaps of lakes and peaks shrink a lot, and every town of 1,000 people or more still gets a point within 3 km.

Measurements

• measurements_current uses one of the about 160 SwissMetNet stations: of the 3 with the smallest distance plus 20 times the height gap, the first that published a temperature in the last hour. Stations are far fewer than forecast points, so the station can still be far from the place and higher or lower.
• Every answer starts with a sentence that says so, such as "The measurements for Braunwald (1290 m) are those of the MeteoSwiss station Elm (958 m), 13.5 km away and 332 m lower."
• It gives the temperature, humidity, dew point, rain and sunshine in the last 10 minutes, wind, gust, direction and sea-level pressure, plus the temperature and pressure change over the last 3 hours and the rain, sunshine and strongest gust of the last hour.
• measurements_extremes lists the first 5 stations by temperature or gust. For the wettest, it lists the first 5 rainy stations and counts the rainy stations in Switzerland and in each canton.
• For the sunniest it groups the stations by their sun in the last 10 minutes with the sunshine terms of MeteoSwiss: mostly sunny (8 to 10 minutes, 80 to 100 %), rather sunny (5 to 7), partly sunny (2 to 4), only a little sunshine (1) and hardly any sunshine (0). MeteoSwiss's ranges meet at 20, 50 and 80 %; a station exactly there counts for the sunnier term, so 2 minutes (20 %) is partly sunny. The cantons come in the order of their stations' average sun.

Forecast Values Are Medians

• Temperature and rain are the median of the many possible outcomes MeteoSwiss calculates, not the most likely value.
• Medians do not add up: when showers are possible but unlikely in any 3-hour block, every block shows 0 mm while the day's median is above 0.
• forecast_daily gives the rain of a day, and forecast_rain the chance and how much a wet hour may bring.

Cache and Downloads

• The cache lives in your OS cache directory, such as ~/Library/Caches/swiss-weather-mcp on macOS or ~/.cache/swiss-weather-mcp on Linux. SWISS_WEATHER_MCP_CACHE_DIR moves it elsewhere.
• The first start downloads the swissNAMES3D archive (about 33 MB) and the swissBOUNDARIES3D GeoPackage (about 37 MB, kept unpacked at about 74 MB).
• MeteoSwiss publishes a new forecast run every hour, each hourly parameter as one file of about 30 to 34 MB for all forecast points. By default the server downloads such a file for every new forecast point and keeps only that point's rows.
• With --cache-all-forecast-points it keeps the whole files, so further places in the same run need no new download. This helps when many places are asked within the hour, at the cost of disk space.

Example Usage with LMStudio

Using the streamable-http transport layer

Configure the mcp.json file in LMStudio:

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

Run the MCP server with the streamable-http transport, as in Running the Server.

Using the stdio transport layer

Configure the mcp.json file in LMStudio. Replace <path-to> so the command is the full path that which swiss-weather-mcp-server prints:

{
  "mcpServers": {
    "swiss_weather_mcp_server": {
      "command": "<path-to>/swiss-weather-mcp-server"
    }
  }
}

Resources

• MeteoSwiss Open Data
• Local Forecast Notebook Examples
• Model Context Protocol
• MCP Server Quickstart

License

Licensed under the Apache License 2.0.