MCP Prefect

🚀 Prefabricated MCP Server

This is an implementation of a Model Context Protocol (MCP) server for Prefect, allowing AI assistants to interact with Prefect through natural language.

🚀 Quick Start

This MCP server enables AI assistants to interact with Prefect via natural language, providing access to various Prefect APIs.

✨ Features

This MCP server provides access to the following Prefect APIs:

• Flow Management: List, retrieve, and delete flows.
• Flow Run Management: Create, monitor, and control flow runs.
• Deployment Management: Manage and schedule deployments.
• Task Run Management: Monitor and control task runs.
• Work Queue Management: Create and manage work queues.
• Block Management: Access block types and documentation.
• Variable Management: Create and manage variables.
• Workspace Management: Get information about workspaces.

📦 Installation

Set the following environment variables:

export PREFECT_API_URL="http://localhost:4200/api"  # The URL of your Prefect API
export PREFECT_API_KEY="your_api_key"               # Your Prefect API key (if using Prefect Cloud)

💻 Usage Examples

Basic Usage

Run the MCP server and Prefect:

docker compose up

Advanced Usage

Once connected, AI assistants can use natural language to help users interact with Prefect. Examples include:

• "Show all my flows"
• "List all failed flow runs from yesterday"
• "Trigger the 'data-processing' deployment"
• "Pause the schedule of the 'daily-reporting' deployment"
• "What was the status of my last ETL flow run?"

🔧 Technical Details

Some endpoints are not yet implemented.

Adding New Features

To add new features to an existing API:

1. Add the function to the corresponding module in the src/mcp_prefect directory.
2. Add the function to the get_all_functions() list in that module.

To add a new API type:

1. Add a new type to APIType (in enums.py).
2. Create a new module in the src/prefect/ directory.
3. Update main.py to include the new API type.

Example usage:

{
  "mcpServers": {
    "mcp-prefect": {
      "command": "mcp-prefect",
      "args": [
        "--transport", "sse"
      ],
      "env": {
        "PYTHONPATH": "/path/to/your/project/directory"
      },
      "cwd": "/path/to/your/project/directory"
    }
  }
}