MCP Weather Xqw

🚀 MCP-weather

An example of an MCP Server for AI Agents to obtain weather forecasts and warning information for US states.

• Original project: mcp-sse

📚 Purpose

This MCP server demonstrates the working mode of an SSE-based MCP server and an independent MCP client using its tools.

🛠️ Operating Environment

This project uses uv to manage related dependencies and the Python runtime environment. If uv is not installed yet, you can follow the installation instructions on the official website.

The following commands are executed in the Ubuntu 24.04 environment. Please modify them according to your operating system:

$ curl -LsSf https://astral.sh/uv/install.sh | sh

Download the source code:

$ git clone https://github.com/erhwenkuo/mcp-weather.git
$ cd mcp-weather
$ uv sync

🚀 Run the Service

Start with uv:

$ uv run server
INFO:     Started server process [15784]
INFO:     Waiting for application startup.
INFO:     Application startup complete.
INFO:     Uvicorn running on http://0.0.0.0:8080 (Press CTRL+C to quit)

Start with Docker

First, build the Docker image:

$ docker build -t mcp-weather .

Start mcp-weather:

$ docker run -d -p 5488:5488 mcp-weather

✅ Verify the Results

First, install Node.js:

# Download and install nvm:
curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.40.2/install.sh | bash

# Source the shell environment variables
\. "$HOME/.nvm/nvm.sh"

# Install Node.js:
nvm install 22

# Verify the Node.js version:
node -v # Should output "v22.14.0".
nvm current # Should output "v22.14.0".

# Verify the npm version:
npm -v # Should output "10.9.2".

Then, start mcp inspector:

$ npx @modelcontextprotocol/inspector
Starting MCP inspector...
Proxy server listening on port 3000

Open your browser and go to http://localhost:3000, then follow the instructions to test the mcp-weather server.

❓ Why Choose SSE

This project uses SSE (Streaming Strategy Engine) as the communication protocol for the following reasons:

1. It supports long connections, suitable for real-time data transmission.
2. It can decouple the sender and receiver, enhancing the scalability of the system.
3. It is suitable for cloud-native environments.

The SSE server runs on localhost:5488 by default, but it can be configured via command-line parameters:

uv run server.py --host <your host> --port <your port>