DEV Community

Mary Luz Chura Ticona
Mary Luz Chura Ticona

Posted on

Building a Weather MCP Server with Python

#python #ai #mcp #tutorial

Building a Weather MCP Server with Python

Introduction

The Model Context Protocol (MCP) is an open standard that allows AI models like Claude to connect with external tools and data sources. In this article, I'll show you how to build a Weather MCP Server using Python that any MCP client can use.

What We'll Build

A simple MCP Server with one tool: get_weather — it receives a city name and returns current temperature, wind speed, and humidity using the Open-Meteo API (free, no API key required).

Requirements

  • Python 3.12+
  • mcp[cli]
  • httpx

Installation 

pip install "mcp[cli]" httpx
Enter fullscreen mode Exit fullscreen mode

The Code 

from mcp.server.fastmcp import FastMCP
import httpx

mcp = FastMCP("Weather MCP Server")

@mcp.tool()
async def get_weather(city: str) -> str:
    """Get current weather for a city."""

    geo_url = f"https://geocoding-api.open-meteo.com/v1/search?name={city}&count=1"
    async with httpx.AsyncClient() as client:
        geo_data = (await client.get(geo_url)).json()

    if not geo_data.get("results"):
        return f"City '{city}' not found."

    r = geo_data["results"][0]
    lat, lon = r["latitude"], r["longitude"]

    weather_url = (
        f"https://api.open-meteo.com/v1/forecast"
        f"?latitude={lat}&longitude={lon}"
        f"&current=temperature_2m,wind_speed_10m,relative_humidity_2m"
        f"&timezone=auto"
    )
    async with httpx.AsyncClient() as client:
        current = (await client.get(weather_url)).json()["current"]

    return (
        f"Weather in {r['name']}, {r['country']}:\n"
        f"Temperature: {current['temperature_2m']}°C\n"
        f"Wind Speed: {current['wind_speed_10m']} km/h\n"
        f"Humidity: {current['relative_humidity_2m']}%"
    )

if __name__ == "__main__":
    mcp.run()
Enter fullscreen mode Exit fullscreen mode

Testing with MCP Inspector

Run the development server:

mcp dev server.py
Enter fullscreen mode Exit fullscreen mode

This opens the MCP Inspector in your browser where you can test the get_weather tool with any city name.

Result

When tested with "Lima", the server returned:

  • Temperature: 19.0°C
  • Wind Speed: 12.2 km/h
  • Humidity: 81%

Repository

The full source code is available on GitHub:
weather-mcp-server

Conclusion

MCP makes it easy to extend AI capabilities with real-world data. With just a few lines of Python, we built a working weather tool that any MCP client like Claude or VSCode can use.

Top comments (1)

Collapse
 
edgar_diegocharaapaza_0 profile image
Edgar Diego CHARA APAZA •

muy útil la forma en que se explica paso a paso cómo construir un servidor MCP con Python. El ejemplo del clima usando Open-Meteo es práctico, sencillo y demuestra muy bien cómo MCP permite conectar modelos de IA con datos reales externos. También me parece interesante que no se necesite una API key, lo que facilita mucho la prueba para quienes están empezando. Buen aporte para entender cómo crear herramientas MCP reutilizables en clientes como Claude o VS Code.