Weather MCP Server 🌦️

A simple, ready-to-use MCP (Model Context Protocol) Server that provides real-time weather information via the OpenWeatherMap API.

Perfect for AI agents, Claude Desktop, TRAE IDE, and any MCP-compatible client.

---

✨ Features

• 🌍 Query current weather by city name
• 🧠 Designed for LLM tool calling (MCP)
• ⚡ Fast startup with uvx
• 🔐 Secure API key handling via environment variables

---

📦 Installation

Option 1: Run directly (recommended)

uvx weather-mcp-server

Option 2: Install from PyPI

pip install weather-mcp-server
weather-mcp-server

---

🔑 Setup

1. Get a free API key from OpenWeatherMap.
2. Set the environment variable:

export OPENWEATHER_API_KEY="your_api_key_here"

On Windows (PowerShell):

$env:OPENWEATHER_API_KEY="your_api_key_here"

---

🛠️ Usage with MCP Clients

Example: TRAE IDE / Claude Desktop

Add this to your MCP configuration:

{
  "mcpServers": {
    "weather": {
      "command": "uvx",
      "args": ["weather-mcp-server"],
      "env": {
        "OPENWEATHER_API_KEY": "your_api_key_here"
      }
    }
  }
}

Then ask your assistant:

“What’s the weather in Beijing?”

---

🧩 Available Tools

query_weather

query_weather(city: str) -> str

Returns a human-readable weather report.

Example output:

Beijing CN 的天气情况：晴，温度：23°C，湿度：45%，风速：3 m/s

---

🚀 Development

Clone and run locally:

git clone https://github.com/yourusername/weather-mcp-server.git
cd weather-mcp-server
uv sync
uv run weather-mcp-server

---

📄 License

MIT License — see LICENSE

---

🙌 Acknowledgements

• Model Context Protocol
• OpenWeatherMap
• FastMCP

---

Made with ☀️ by [Your Name]

📌 **记得修改两处占位符**：
1. `your_api_key_here` → 提醒用户替换（不用真填）
2. `yourusername` → 你的 GitHub 用户名
3. `[Your Name]` → 你的名字

---

## 四、现在你的项目结构长这样 ✅

weather-mcp-server/ ├── pyproject.toml ├── README.md ├── LICENSE ├── .python-version └── src/ └── weather_mcp/ ├── init.py └── server.py

---

## 五、下一步你可以做什么

1. `uv build` → 构建 wheel
2. `uv publish` → 发布到 PyPI
3. 在 TRAE / Claude Desktop 里用 `uvx weather-mcp-server` 体验“一键运行”

如果你愿意，下一步我可以帮你：
- ✅ 检查 `pyproject.toml` 是否 100% 符合 PyPI 要求
- ✅ 帮你写 `.gitignore`
- ✅ 模拟一次“发布前的最终检查清单”
- ✅ 教你如何给 MCP Server 加版本号和 CHANGELOG

你想继续哪一步？ 😊