MCP ServerSTDIOOfficialv0.9.0

Abhishekbhakat Airflow MCP Server

Manage and monitor Apache Airflow DAGs, tasks, and runs directly from your AI chat. Connect to any Airflow instance via API to trigger workflows, check execution status, and inspect logs.

io.github.abhishekbhakat/airflow-mcp-server

Hosted URL

Local install

Transport

STDIO

Auth

4 required env vars

What the Abhishekbhakat Airflow MCP server does

How models use it and what it is built for.

Manage and monitor Apache Airflow DAGs, tasks, and runs directly from your AI chat. Connect to any Airflow instance via API to trigger workflows, check execution status, and inspect logs.

Connect to Abhishekbhakat Airflow

Local install — runs as a subprocess.

uvx airflow-mcp-server==0.9.0

Environment variables

Configuration this server reads at startup.

  • AIRFLOW_BASE_URLRequired

    The base URL for the Airflow API (e.g., http://localhost:8080)

  • AUTH_TOKENRequiredSecret

    The JWT authentication token for Airflow API access

  • AIRFLOW_BASE_URLRequired

    The base URL for the Airflow API (e.g., http://localhost:8080)

  • AUTH_TOKENRequiredSecret

    The JWT authentication token for Airflow API access

Resources

Where to find authoritative docs and source for Abhishekbhakat Airflow.

Example prompts for Abhishekbhakat Airflow

Paste any of these into Agent Studio after connecting Abhishekbhakat Airflow.

  • List all DAGs in my Airflow instance and show their recent run status
  • Trigger a DAG run for my data_pipeline DAG with these parameters
  • Show me the task logs from the last failed run of my ETL workflow
  • What is the current status of all running tasks across my Airflow cluster

Abhishekbhakat Airflow MCP server — FAQ

Common questions about connecting and running Abhishekbhakat Airflow.

  • What do I need to connect this MCP server to Airflow?

    You need two environment variables: AIRFLOW_BASE_URL (the URL to your Airflow instance, e.g., http://localhost:8080) and AUTH_TOKEN (a valid JWT authentication token for the Airflow API). Both are required to establish a connection.

  • How do I generate a JWT token for Airflow API authentication?

    JWT tokens are generated by your Airflow instance. Consult your Airflow admin or documentation for your deployment method (Docker, Kubernetes, standalone) to obtain a valid token with appropriate API permissions.

  • Can I use this server with Airflow running on a remote server?

    Yes. Set AIRFLOW_BASE_URL to the remote Airflow instance's URL (e.g., https://airflow.company.com) and provide a valid AUTH_TOKEN. Ensure network connectivity and that the token has permission to access the remote API.

  • What Airflow operations can I perform through this MCP server?

    Based on the environment variables and design, you can query DAGs, check task and run status, view logs, and trigger DAG runs. Refer to the Airflow REST API documentation for the full scope of supported operations.

  • What happens if my AUTH_TOKEN expires?

    API calls will fail with authentication errors. You will need to generate a new JWT token and update the AUTH_TOKEN environment variable to restore connectivity.

Skip the local setup — run MCP in your browser

MCP Playground runs 10,000+ hosted MCP servers — GitHub, Linear, Notion, Stripe, Sentry and more — across Claude, GPT, Gemini, DeepSeek and 30+ AI models. Compare model answers side-by-side, save agent presets, share runs. Zero install.

Open Agent Studio

Related servers

More on MCP Playground