🌦️ Claude Weather Agent (MCP-Enabled)
This project enables Claude to interact with live weather data (current + forecast) using the OpenWeather API via an MCP (Multi-Command Protocol) server.
It includes:
- FastMCP server using Python
- Weather tools (
get_weatherandget_forecast) - Environment variable-based API security
- Easy integration with
claude_desktop_config.py
🚀 How to Run the MCP Server
- Clone this repo and navigate into it:
git clone https://github.com/ChandekarDhruvin/claude-openweather-mcp.git
cd claude-weather-agent/weather
- Install dependencies:
pip install -r requirements.txt
- Create a
.envfile in the same directory asweather.py:
OPENWEATHER_API_KEY=your_openweather_api_key
- Start the MCP server manually (for testing):
python weather.py
🧠 Tools Exposed to Claude
| Tool | Parameters | Description |
|---|---|---|
get_weather |
city: str, country: str = "IN" |
Returns current weather conditions. |
get_forecast |
city: str, country: str = "IN" |
Provides a 5-day forecast (around noon). |
🛠️ Claude Integration: claude_desktop_config.py
Add the following to your Claude config to auto-start the MCP server:
{
"mcpServers": {
"weather": {
"command": "C:\\\\path\\\\to\\\\uv.exe",
"args": [
"--directory",
"C:\\\\path\\\\to\\\\weather\\\\folder",
"run",
"C:\\\\path\\\\to\\\\weather\\\\weather.py"
]
}
}
}
📝 Note: You must adjust the path to match your local Python virtual environment and directory structure.
📂 Project Structure
weather/
├── weather.py # Main FastMCP server with weather tools
├── .env # Contains the OpenWeather API key
├── requirements.txt # Dependencies
└── README.md # You're reading it!
🔒 Environment Variables
Using python-dotenv, we securely load the OpenWeather API key from a .env file.
from dotenv import load_dotenv
import os
load_dotenv()
OPENWEATHER_API_KEY = os.getenv("OPENWEATHER_API_KEY")
✅ Requirements
Add this in your requirements.txt:
httpx
python-dotenv
fastmcp
User: What's the weather like in Mumbai today?
Claude:
Current weather in Mumbai, IN:
Condition: Clear (clear sky)
Temperature: 32°C (Feels like 35°C)
Humidity: 58%
Wind Speed: 3.5 m/s
Reference
https://modelcontextprotocol.io/
Recommend MCP Servers 💡
@burtthecoder/mcp-shodan
MCP server for querying the Shodan API
lisply-mcp
A Model Context Protocol (MCP) server that connects Large Language Models (LLMs) to Lisp-based development and runtime environments, enabling AI-assisted symbolic programming via the lightweight Lisply protocol.
Kashyap-AI-ML-Solutions/webex-messaging-mcp-server
A Model Context Protocol (MCP) server that provides AI assistants with comprehensive access to Cisco Webex messaging capabilities.
basic-memory
Basic Memory is a local-first knowledge base that integrates with Large Language Models (LLMs) via the Model Context Protocol (MCP), allowing AI assistants to read and write to your local Markdown files, enabling persistent memory for conversations and structured knowledge management.
mcp_server_lark
A Model Context Protocol server for Lark(Feishu) sheet, message, doc and etc.
review-toolkit-mcp
A Model Context Protocol (MCP) service designed to assist AI agents in managing and tracking code review sessions, including file tracking, review status, agent review storage, and report generation.