agri-mcp
Fetches live agricultural weather data via OpenWeather API, providing current conditions and multi-day forecasts for any city.
README
agri-mcp
An MCP (Model Context Protocol) server that gives Claude access to live agricultural weather data via the OpenWeather API. Designed to run as a local tool server for Claude Desktop.
Features
- get_weather — Get current weather conditions (temperature, description, humidity) for any city
- get_forecast — Get a multi-day weather forecast for any city
Requirements
- Python 3.10+
- An OpenWeather API key (free tier works)
Setup
1. Clone / create the project folder
mkdir agri-mcp
cd agri-mcp
2. (Optional) Create a virtual environment
python -m venv venv
venv\Scripts\activate
3. Install dependencies
pip install mcp httpx python-dotenv
4. Add your OpenWeather API key
Create a .env file in the project root (only needed for local/standalone testing — Claude Desktop passes env vars directly from its config):
OPENWEATHER_API_KEY=your_api_key_here
5. Test the server standalone
python main.py
The server communicates over stdio, so it won't print anything visible — if it doesn't immediately crash, it's working. Press Ctrl+C to stop it.
Connecting to Claude Desktop
Edit your Claude Desktop config file:
notepad "%APPDATA%\Claude\claude_desktop_config.json"
Add the following (adjust the path to match where main.py actually lives on your machine):
{
"mcpServers": {
"agri-data": {
"command": "python",
"args": [
"C:\\Users\\YOUR_USERNAME\\PycharmProjects\\PythonProject\\agri-mcp\\main.py"
],
"env": {
"OPENWEATHER_API_KEY": "your_api_key_here"
}
}
}
}
Then fully quit Claude Desktop (not just close the window — end the process from Task Manager or system tray) and reopen it. Check Settings → Developer (or the MCP servers panel) to confirm agri-data shows as connected.
Project Structure
agri-mcp/
├── main.py # MCP server entry point and tool definitions
├── .env # Local environment variables (not committed)
└── README.md
Troubleshooting
| Issue | Fix |
|---|---|
Server disconnected in Claude Desktop |
Check the config JSON structure — command, args, and env must be nested inside mcpServers.agri-data, not at the top level |
python not recognized |
Disable the Python App Execution Alias in Windows Settings, or use the full path to python.exe |
| Old config still showing after edit | Fully quit Claude Desktop (Task Manager → End Task) and reopen — closing the window alone doesn't reload the config |
OPENWEATHER_API_KEY is not set error |
Confirm the key is present in both the config's env block and/or your local .env file |
Security Note
Do not commit your .env file or claude_desktop_config.json with a real API key to version control. If a key is ever accidentally exposed (e.g. shared in a screenshot or chat), regenerate it from your OpenWeather account dashboard.
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。