Apache Airflow MCP

What is Apache Airflow MCP?

Apache Airflow MCP is a Model Context Protocol server for Apache Airflow, the most popular open-source workflow orchestration platform. Airflow schedules and monitors workflows as directed acyclic graphs (DAGs) of tasks.

DAG Management Intelligence

AI models can monitor DAG runs, identify bottlenecks, debug failed tasks, and help optimize pipeline performance. Instead of navigating the Airflow UI, ask your AI assistant about pipeline status.

Task Debugging

When tasks fail, Airflow MCP provides access to task logs, execution context, and retry history. AI models can analyze failure patterns and suggest fixes based on common Airflow issues.

Configuration

{
  "mcpServers": {
    "airflow": {
      "command": "npx",
      "args": ["airflow-mcp"],
      "env": {
        "AIRFLOW_API_URL": "http://localhost:8080/api/v1",
        "AIRFLOW_USERNAME": "admin",
        "AIRFLOW_PASSWORD": "your_password"
      }
    }
  }
}

Use Cases

Airflow MCP serves data engineers managing complex ETL pipelines, platform teams monitoring hundreds of DAGs, and organizations needing AI-powered pipeline debugging and optimization.