Dolphinscheduler MCP

🚀 Apache DolphinScheduler MCP Server

The Model Context Protocol (MCP) server of Apache DolphinScheduler provides a FastMCP-based service, exposing DolphinScheduler's REST API as a series of tools for AI agents to interact with through a standardized protocol. This server acts as a bridge between AI models and DolphinScheduler, enabling AI-driven workflow management.

🚀 Quick Start

The MCP server of Apache DolphinScheduler serves as a crucial link between AI models and DolphinScheduler, facilitating AI-driven workflow management.

✨ Features

• Comprehensive coverage of DolphinScheduler's functionality.
• Standard tool interfaces compliant with the Model Context Protocol.
• Easy configuration via environment variables or command-line parameters.
• Comprehensive tool documentation.

📦 Installation

pip install dolphinscheduler-mcp

📚 Documentation

Configuration

Environment Variables

• DOLPHINSCHEDULER_API_URL: The URL of the DolphinScheduler API (Default: http://localhost:12345/dolphinscheduler).
• DOLPHINSCHEDULER_API_KEY: The API token for authentication with the DolphinScheduler API.
• DOLPHINSCHEDULER_MCP_HOST: The host address to which the MCP server binds (Default: 0.0.0.0).
• DOLPHINSCHEDULER_MCP_PORT: The port to which the MCP server binds (Default: 8089).
• DOLPHINSCHEDULER_MCP_LOG_LEVEL: The logging level (Default: INFO).

Usage

Command Line

Start the server using the command-line interface:

ds-mcp --host 0.0.0.0 --port 8089

Python API

from dolphinscheduler_mcp.server import run_server

# Start the server
run_server(host="0.0.0.0", port=8089)

Available Tools

The tools provided by the DolphinScheduler MCP server include:

• Project management
• Process definition management
• Process instance management
• Task definition management
• Scheduling management
• Resource management
• Data source management
• Alarm group management
• Alarm plugin management
• Worker group management
• Tenant management
• User management
• System status monitoring

💻 Usage Examples

Basic Usage

from mcp_client import MCPClient

# Connect to the MCP server
client = MCPClient("http://localhost:8089/mcp")

# Get the project list
response = await client.invoke_tool("get-project-list")

# Create a new project
response = await client.invoke_tool(
    "create-project", 
    {"name": "My AI Project", "description": "Project created by AI"}
)

📄 License

Apache License 2.0