# Weather MCP Server

A mini project demonstrating how to build and use a Model Context Protocol (MCP) server with Python.

## Features

- Weather Alerts Tool
- Weather Forecast Tool
- SSE Transport
- STDIO Transport
- MCP Client Integration
- Docker Support
- Async Python Implementation

## Project Structure

```
mcpcrashcourse/
│
├── server/
│   └── mcpserver/
│       ├── server.py
│       ├── client-sse.py
│       ├── client-stdio.py
│       ├── Dockerfile
│       └── requirements.txt
│
├── weather.py
├── weather.json
├── client.py
├── README.md
├── pyproject.toml
└── uv.lock
```

## Tools

### get_alerts

Returns active weather alerts for a US state.

Example:

```python
get_alerts("CA")
```

### get_forecast

Returns forecast information based on latitude and longitude.

Example:

```python
get_forecast(
    latitude=37.7749,
    longitude=-122.4194
)
```

## Transports

### STDIO

Used for local MCP communication.

### SSE

Used for remote communication through HTTP.

## Running the Server

### STDIO

```bash
uv run server/mcpserver/server.py
```

### SSE

Configure transport as:

```python
mcp.run(transport="sse")
```

Then:

```bash
uv run server/mcpserver/server.py
```

## Running Clients

### SSE Client

```bash
uv run server/mcpserver/client-sse.py
```

### STDIO Client

```bash
uv run server/mcpserver/client-stdio.py
```

## Docker

Build:

```bash
docker build -t weather-mcp .
```

Run:

```bash
docker run weather-mcp
```

## Technologies Used

- Python
- MCP
- FastMCP
- HTTPX
- AsyncIO
- Docker
- SSE
- STDIO

## Learning Outcomes

- Building MCP tools
- Creating MCP servers
- Working with MCP clients
- Using SSE transport
- Using STDIO transport
- Tool invocation
- Agent-tool communication
- Dockerizing MCP applications