Email Creation with AI Agents | Beefree SDK + PydanticAI MCP Showcase

Are you exploring agentic email design? We believe the future of content design means having AI agents and humans work seamlessly together to create content — and this simple demo app shows you how to create a design experience that does just that.

It connects the Beefree SDK (the most intuitive drag-and-drop email editor in the market) with a PydanticAI agent using the Model Context Protocol (MCP), and supports Anthropic, OpenAI, and Gemini as the LLM provider.

Here's what you can achieve:

• AI-Powered Email Design: Use natural language to create and modify email templates
• WYSIWYG Editor: Allow humans to easily edit drafts created by AI in the intuitive, drag-and-drop editor provided by the Beefree SDK

What powers this experience:

• Real-time Streaming: WebSocket-based streaming chat interface
• MCP Integration: Direct connection to Beefree's Streamble HTTP MCP Server.
• Modern Tooling: Built with uv for fast dependency management

Prerequisites

• Python 3.13+
• uv for package management (recommended)
• Beefree SDK account (If you already have an account, you can get your credentials here. If you're new to Beefree SDK, you can sign up for a free account.)
• Anthropic, OpenAI, or Gemini API key
• Beta access. Learn how to request access to the beta in our Beefree SDK MCP Server (Beta) documentation.

Quick start

1. Clone and install

# Clone the repository
git clone <repository-url>
cd beefree_mcp_example

# Install with uv
uv sync

2. Configure environment

Create a .env file and add your credentials. Pick the example that matches your provider. Anthropic:

AI_PROVIDER=anthropic
BEEFREE_CLIENT_ID=your_beefree_client_id
BEEFREE_CLIENT_SECRET=your_beefree_client_secret
BEEFREE_MCP_API_KEY=your_beefree_mcp_api_key
ANTHROPIC_API_KEY=your_anthropic_api_key
LLM_MODEL=claude-opus-4-5

OpenAI:

AI_PROVIDER=openai
BEEFREE_CLIENT_ID=your_beefree_client_id
BEEFREE_CLIENT_SECRET=your_beefree_client_secret
BEEFREE_MCP_API_KEY=your_beefree_mcp_api_key
OPENAI_API_KEY=your_openai_api_key
LLM_MODEL=gpt-5.2

Gemini:

AI_PROVIDER=gemini
BEEFREE_CLIENT_ID=your_beefree_client_id
BEEFREE_CLIENT_SECRET=your_beefree_client_secret
BEEFREE_MCP_API_KEY=your_beefree_mcp_api_key
GEMINI_API_KEY=your_gemini_api_key
LLM_MODEL=gemini-2.5-pro

3. Run the application

# With uv
uv run python main.py

The application will be available at http://localhost:8000

How it works

1. User Input: The user types a natural language request in the chat interface.
2. Agent Run: The PydanticAI agent runs with conversation history and editor context.
3. MCP Execution: Direct HTTP calls to Beefree's MCP server modify the email template.
4. Progress Updates: The UI shows step-by-step progress messages.
5. Editor Integration: Changes are reflected in the Beefree editor.

Resources

About the Beefree SDK

• Beefree SDK Documentation
• Beefree SDK MCP Server Documentation

About PydanticAI:

• PydanticAI Documentation
• PydanticAI MCP Integration

More resources:

• MCP tool list for this demo: mcp-tool-list.md
• What's a Model Context Protocol?
• uv Documentation