MCP Airflow API

What it does

MCP Airflow API bridges the gap between LLMs and Apache Airflow by transforming complex REST API operations into natural language tools. It allows data engineers to monitor DAGs, trigger workflows, and manage cluster health without manually interacting with the Airflow UI or writing curl commands.

Tools

• DAG Management: Trigger, pause, resume, and list DAGs with advanced filtering.
• Task Monitoring: Retrieve task instance logs, track execution states, and analyze durations.
• Cluster Health: Check version info and overall system health.
• XCom & Variable Management: Read and manage Airflow variables and XCom entries.
• Asset Management (v2): Manage data assets and events for data-aware scheduling in Airflow 3.0+.

Installation

For Claude Desktop, use the following configuration:

{
  "mcpServers": {
    "mcp-airflow-api": {
      "command": "uvx",
      "args": ["--python", "3.12", "mcp-airflow-api"],
      "env": {
        "AIRFLOW_API_VERSION": "v2",
        "AIRFLOW_API_BASE_URL": "http://localhost:8080/api",
        "AIRFLOW_API_USERNAME": "airflow",
        "AIRFLOW_API_PASSWORD": "airflow"
      }
    }
  }
}

Supported hosts

• claude