# MCP Weather Project

A Model Context Protocol (MCP) server implementation that provides weather alerts via the National Weather Service (NWS) API. This project includes both a FastMCP server and a LangChain-based client with memory capabilities.

## Features

- **Weather Alerts**: Fetch active weather alerts for any US state using the NWS API.
- **Echo Resource**: A simple resource that echoes back messages.
- **Greeting Prompt**: A customizable greeting prompt generator.
- **Interactive Client**: A CLI-based chat client powered by Groq's Llama 3.3 model with conversation memory.

## Prerequisites

- Python 3.12 or higher
- [uv](https://github.com/astral-sh/uv) (recommended for dependency management)
- A [Groq](https://console.groq.com/) API Key for the client.

## Installation

1. **Clone the repository:**
   ```bash
   git clone <repository_url>
   cd mcpfile
   ```

2. **Install dependencies:**
   Using `uv` (recommended):
   ```bash
   uv sync
   ```
   Or using pip:
   ```bash
   pip install -r requirements.txt
   ```
   *(Note: You may need to generate a `requirements.txt` from `pyproject.toml` if not using `uv`)*

3. **Set up Environment Variables:**
   Create a `.env` file in the root directory and add your Groq API key:
   ```bash
   GROQ_API_KEY=your_groq_api_key_here
   ```

## Usage

### Running the Entry Point

The `main.py` is a simple entry point script that prints a welcome message.
```bash
uv run main.py
# OR
python main.py
```

### Running the interactive Client

The client connects to the weather server and allows you to interact with it using natural language.

1. Ensure the server configuration in `server/weather.json` is correct (it points to `server/weather.py`).
2. Run the client:
   ```bash
   uv run server/client.py
   # OR if using a virtual environment directly:
   # python server/client.py
   ```

3. **Example Interaction:**
   ```text
   You: Check weather alerts for TX
   Assistant: Checking weather alerts for Texas...
   [Agent responds with alerts]
   ```

### Running the MCP Server Standalone

You can run the MCP server directly using `uv`. This is useful for inspection or debugging with the MCP Inspector.

```bash
uv run --with mcp[cli] mcp run server/weather.py
```

### Configuration Verification

Ensure that `server/weather.json` points to the correct absolute path of your `server/weather.py` file.
```json
{
  "mcpServers": {
    "weather": {
      "command": "uv",
      "args": [
        "run",
        "--with",
        "mcp[cli]",
        "mcp",
        "run",
        "/your/absolute/path/to/mcp/mcpfile/server/weather.py" 
      ]
    }
  }
}
```

### Project Structure

- **`server/weather.py`**: The main MCP server implementation using `FastMCP`. Defines tools (`get_alerts`), resources, and prompts.
- **`server/client.py`**: An MCP client implementation using `LangChain` and `ChatGroq`. Handles the interactive chat session.
- **`server/weather.json`**: Configuration file for the MCP client to locate the server.
- **`main.py`**: Simple entry point script.
- **`pyproject.toml`**: Project configuration and dependencies.

## Tools Available

- `get_alerts(state: str)`: Get active weather alerts for a US state (e.g., "CA", "NY").

## Resources

- `echo://{message}`: Echoes a message.

## Prompts

- `greet_user(name: str, style: str)`: Generates a greeting in a specified style ("friendly", "formal", or "casual").