Weather Alerts MCP Server
Provides real-time weather alerts for US states using the National Weather Service API, enabling users to fetch and view active weather warnings through natural language queries.
README
MCP Server Beginner example
Project Overview
It is a Python application that provides real-time weather alerts for US states using the National Weather Service (NWS) API. It leverages the mcp framework (with FastMCP) and is designed for modern async Python workflows. The project is fully managed with uv for dependency management and task running.
Features
- Weather Alerts Tool: Fetches and formats active weather alerts for any US state.
- Echo Resource: Simple echo endpoint for testing.
- Async & Modern: Uses Python 3.12+ and async/await patterns.
- MCP Inspector: Visual tool to inspect and test your MCP tools and resources.
Project Structure
.
├── main.py # Simple entry point (prints a message)
├── pyproject.toml # Project metadata and dependencies
├── uv.lock # Locked dependencies for reproducible installs
├── server/
│ └── weather.py # Main server logic and weather tool/resource
├── .venv/ # (optional) Virtual environment
├── .python-version # Python version pin (3.12)
├── .gitignore # Standard Python ignores
└── README.md # Project documentation
Prerequisites
- Python 3.12+
- uv (install with
pip install uvif not already installed)
Setup & Installation
-
Clone the repository:
git clone <your-repo-url> cd mcpproject -
Install dependencies using uv:
uv pip install -r uv.lockOr, if you want to install from
pyproject.toml:uv pip install
Running the Project
1. Run the Main Script
To run the basic entry point:
uv run python main.py
2. Develop & Test the Weather Tool
To start the MCP development server for the weather tool:
uv run mcp dev server/weather.py
This will also open MCP Inspector in your browser—a visual tool to explore and test your MCP tools and resources interactively.
3. Install the Weather Tool
To install the weather tool for use as an MCP tool:
uv run mcp install server/weather.py
Usage Example
With the dev server running, you can use the MCP Inspector or an MCP-compatible client to call the get_alerts tool:
- Tool name:
get_alerts - Argument: Two-letter US state code (e.g.,
CA,NY)
Example (from CLI):
mcp call get_alerts --state=CA
Using with Claude Desktop
You can also use the weather alerts tool in Claude Desktop, as shown below:

Weather alerts for California displayed in Claude Desktop using the get_alerts tool.
Claude Desktop Configuration
To call this MCP server from Claude Desktop, you need to update your claude_desktop_config.json file to include the server
For example:
{
{
"mcpServers": {
"weather": {
"command": "C:\\Users\\shubhechhadey\\.local\\bin\\uv.EXE",
"args": [
"run",
"--with",
"mcp[cli]",
"mcp",
"run",
"C:\\Users\\shubhechhadey\\projects\\mcpproject\\server\\weather.py"
]
}
}
}
}
Attached is the screenshot: 
Notes
- All commands use
uv
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。