Global Weather Server

A Model Context Protocol (MCP) server that gives AI assistants access to weather data without API keys.

• US weather alerts from the National Weather Service API
• US point forecasts from the National Weather Service
• Global 5-day forecasts from Open-Meteo, including India

Requirements

Requirement	Version
Python	3.14+
uv	Latest
Claude Desktop or another MCP client	Latest

Check your local versions:

python --version
uv --version

Installation

git clone https://github.com/YOUR-USERNAME/global-weather-mcp-server.git
cd global-weather-mcp-server
uv sync

No API keys are required.

Connect to Claude Desktop on Windows

For the Microsoft Store/package install of Claude Desktop, open this config file:

%LOCALAPPDATA%\Packages\Claude_pzs8sxrjxfjjc\LocalCache\Roaming\Claude\claude_desktop_config.json

Add only this server entry inside the mcpServers object:

{
  "mcpServers": {
    "global-weather": {
      "command": "uv",
      "args": [
        "--directory",
        "C:\\path\\to\\weather-mcp",
        "run",
        "global_weather_mcp_server.py"
      ]
    }
  }
}

If the config file already has other MCP servers, keep them and add only the global-weather block. Restart Claude Desktop after saving the file. The server should expose these tools:

• get_alerts
• get_forecast
• get_forecast_global

Tools

get_alerts

Get active weather alerts for a US state.

Example:

What are the active weather alerts in California?

get_forecast

Get a detailed National Weather Service forecast for a US location by latitude and longitude.

Example:

What's the weather forecast for latitude 40.71, longitude -74.01?

get_forecast_global

Get current conditions and a 5-day forecast for any location worldwide by latitude and longitude.

Examples:

What's the weather forecast for latitude 19.0760, longitude 72.8777?
Get the weather for latitude 28.6139, longitude 77.2090.

Useful Indian city coordinates:

City	Latitude	Longitude
New Delhi	28.6139	77.2090
Mumbai	19.0760	72.8777
Bangalore	12.9716	77.5946
Chennai	13.0827	80.2707
Kolkata	22.5726	88.3639
Hyderabad	17.3850	78.4867
Pune	18.5204	73.8567

Test Locally

Run the MCP server directly:

uv run global_weather_mcp_server.py

If it starts and waits silently, the server is ready for an MCP client.

Project Structure

global-weather-mcp-server/
|-- global_weather_mcp_server.py
|-- pyproject.toml
|-- uv.lock
|-- README.md
`-- .gitignore