pydantic mcp agent

# Pydantic MCP Agent Example

This project demonstrates a minimal setup for integrating Pydantic AI agents with MCP servers, following the pattern from ottomator-agents/pydantic-ai-mcp-agent.

## Setup

1. **Create a virtual environment with uv**
    ```bash
    uv venv .venv
    source .venv/bin/activate
    uv pip install --upgrade pip
    pip install pydantic-ai mcp
    ```

2. **Copy and configure files**
    - `mcp_client.py`: MCP client logic (copied from the reference repo)
    - `pydantic_mcp_agent.py`: Minimal agent script
    - `mcp_config.json`: MCP server config (edit as needed)

3. **Set up API keys**
    - If needed, create a `.env` file with your keys, following the `.env.example` pattern from the reference repo.

4. **Run your agent**
    - See `pydantic_mcp_agent.py` for usage. You will need to provide your model and ensure your MCP config is correct.

## Reference
See the original [ottomator-agents/pydantic-ai-mcp-agent](../ottomator-agents/pydantic-ai-mcp-agent) for full details and advanced usage.