seatunnel

SeaTunnel MCP Server provides a bridge between large language model interfaces and SeaTunnel capabilities. It exposes tools to manage SeaTunnel connections, submit and monitor jobs, and retrieve system information and statistics. You can interact with these tools via MCP-compatible clients or through the MCP Inspector to test and explore endpoints. The server runs as a Python module and can be started with the command: python -m src.seatunnel_mcp. When used with Claude Desktop, you can configure the MCP server entry to point to this module so Claude can invoke the available actions. The available tool categories include Connection Management, Job Management, and System Monitoring, enabling you to view and update connection settings, submit single or multiple jobs, stop jobs, fetch job details, and obtain cluster overview. For testing and development, you can also use the MCP Inspector to send requests to the Python module, ensuring the actions and payloads conform to the MCP specification.

Typical workflows include configuring the SeaTunnel REST API endpoint, submitting a new SeaTunnel job with its configuration, and monitoring running or finished jobs. The server also supports dynamic configuration and provides commands for viewing current connection settings and updating them to point to different SeaTunnel instances. For local experimentation, ensure a SeaTunnel instance is running and accessible at the configured API URL (default http://localhost:8090).

Prerequisites:

• Python 3.12 or newer
• SeaTunnel instance running and accessible via REST API
• Optional: Node.js for testing with MCP Inspector

Installation steps:

1. Clone the repository git clone <repository_url> cd seatunnel-mcp

2. Create a Python virtual environment and activate it python -m venv .venv source .venv/bin/activate # On Windows: .venv\Scripts\activate

3. Install the MCP server in editable mode pip install -e .

4. Run the MCP server python -m src.seatunnel_mcp

5. (Optional) Install and run MCP Inspector for testing npm i -g @modelcontextprotocol/inspector npx @modelcontextprotocol/inspector python -m src.seatunnel_mcp

Notes:

• Ensure the SeaTunnel REST API URL and API key (if required) are correctly configured via environment variables or in the server config.
• If you modify dependencies, you may need to reactivate the virtual environment or reinstall.
• The installation assumes you are running from the project root where the src.seatunnel_mcp module exists.

Environment variables:

• SEATUNNEL_API_URL: Base URL for SeaTunnel REST API (default http://localhost:8090)
• SEATUNNEL_API_KEY: Optional API key for SeaTunnel REST API

Common issues:

• If the SeaTunnel REST API is not reachable, verify network connectivity and the API URL configured in SEATUNNEL_API_URL.
• Ensure Python 3.12+ is installed and the virtual environment is active when running the MCP server.
• When using Claude Desktop or MCP Inspector, ensure the command and arguments match the module path and entry point shown in the usage examples.

Configuration tips:

• You can override the SeaTunnel API URL by setting SEATUNNEL_API_URL to point to a different SeaTunnel instance.
• If you need to connect to a new SeaTunnel cluster, update the environment variables or the CLI configuration accordingly before invoking job operations.