Weather API MCP
# 🌦️ Weather API MCP
A zero-configuration Model Context Protocol (MCP) weather server for Claude and other MCP clients. It provides current conditions and forecasts using the free Open-Meteo APIs—no weather API key required.
## Features
- Current weather for locations worldwide
- Forecasts for 1–7 days
- Celsius or Fahrenheit temperatures
- No weather API key required
- Python implementation using the MCP Python SDK
## Requirements
- Python 3.10+
- [`uv`](https://docs.astral.sh/uv/)
## Install and run
```bash
uv sync
uv run weather-mcp
```
The server uses stdio transport for MCP clients. To install it into the active Python environment instead, run `pip install -e .` and then `weather-mcp`.
## Configure Claude Desktop
After installing the project with `uv sync`, add the following to your Claude Desktop MCP configuration, replacing `/absolute/path/to/WEATHER_API_MCP` with this repository's directory:
```json
{
"mcpServers": {
"weather": {
"command": "uv",
"args": [
"--directory",
"/absolute/path/to/WEATHER_API_MCP",
"run",
"weather-mcp"
]
}
}
}
```
## Tools
### `get_weather`
Parameters: `location` (string), `unit` (`celsius` or `fahrenheit`, optional; defaults to Celsius).
Returns the resolved location, temperature, apparent temperature, weather condition, humidity, wind speed, and pressure.
### `get_forecast`
Parameters: `location` (string), `days` (integer from 1 to 7, optional; defaults to 5), and `unit` (`celsius` or `fahrenheit`, optional; defaults to Celsius).
Returns daily high and low temperatures, conditions, and precipitation probabilities.
## Data source
Location lookup and weather data are provided by [Open-Meteo](https://open-meteo.com/). Internet access is required; no account or API key is needed.
TDQS
Scored across 2 tools
get_weather and get_forecast have clearly distinct purposes: current conditions versus daily forecast. An agent can easily choose the correct tool based on whether it needs present weather or a multi-day outlook.
Both tools follow a consistent get_<noun> pattern. The naming is predictable and directly reflects the returned data.
Two tools are thin for a weather API surface, though they cover the most common current-weather and forecast use cases. Additional tools for hourly forecasts, alerts, or location lookup would make the set feel more complete.
The core weather workflow—current conditions and 1–7 day forecasts—is covered without dead ends. Minor gaps remain around hourly forecasts, severe weather alerts, and location resolution/geocoding.