Weather

You will build a lightweight MCP Server in Python that exposes real-time weather data from the Open-Meteo API as a tool. This lets language models request current weather data for a location and receive structured results that they can incorporate into conversations or workflows.

How to use

Use an MCP client to connect to the weather MCP server and call the get_current_weather tool with a latitude and longitude. The tool retrieves current weather data for the specified location and returns the raw data for the model to interpret or further process. You can extend the server later with additional tools like get_forecast for forecasts or get_location for geocoding to improve location accuracy.

How to install

Prerequisites you need before building the server are:

• A code editor (for example, VS Code)
• uv a Rust-based Python package manager
• Python (for running the MCP server and dependencies)
• A code editor to review and customize the code
• A tool to run commands in your terminal (like zsh, bash, or PowerShell)

Install uv on your system:

• macOS: open a terminal and run brew install uv
• Windows: open a terminal and run winget install --id=astral-sh.uv -e

Create and prepare your project folder:

• Create a new folder mkdir mcp-server-weather
• Navigate into the folder cd mcp-server-weather
• Initialize a new uv project uv init
• Create a virtual environment uv venv
• Activate the virtual environment source .venv/bin/activate (on Windows use .
envin o_activate equivalent)
• Install the Python MCP SDK with CLI extension and dependencies uv add "mcp[cli]" httpx

Create the server scaffold. In your project folder create a file named server.py (or reuse main.py if you prefer). Add the following scaffold to initialize and run the server with stdio transport:

from typing import Any
import httpx
from mcp.server.fastmcp import FastMCP

# Initialize FastMCP server
mcp = FastMCP("weather")

### The rest of your code goes between here...

### ... and here.

if __name__ == "__main__":
    # Initialize and run the server
    mcp.run(transport='stdio')

Run the MCP server in development mode

To run the server in development mode and start the MCP Inspector for testing, use the following command:

mcp dev server.py

Available tools

get_current_weather

A tool that fetches current weather data for a given latitude and longitude using the Open-Meteo API. It returns the full data payload for the model to process.