MCP Server for Apache Airflow

You run an MCP server that lets AI assistants interact with Apache Airflow. It exposes DAG management, DAG run monitoring, task instance details, and logs, so you can control workflows and debug issues directly from your AI workflows or chat interfaces.

How to use

Connect your MCP client to the Airflow MCP server either in stdio (local development) or via HTTP (cloud deployment). Once connected, you can list DAGs, view details, trigger runs, pause or unpause DAGs, and monitor DAG runs and task instances. Retrieve task logs, DAG run logs, and tail ongoing DAG runs to observe real-time activity. Use natural language requests to perform common Airflow actions and inspect results in your preferred interface.

How to install

Prerequisites: you need Node.js and npm installed on your machine.

1. Clone the MCP Airflow server repository.

2. Install dependencies.

3. Build the project.

4. Start the server in the mode you need.

git clone https://github.com/tomnagengast/mcp-server-airflow.git
cd mcp-server-airflow
npm install
npm run build

# For stdio mode (Claude Desktop)
npm start

# For HTTP mode (cloud deployment)
npm run start:http
"}]} ,{

Configuration and operation notes

You authenticate with Airflow using either an API token or basic credentials. Provide the base URL of your Airflow instance and the appropriate credentials via environment variables when you run the server. The MCP server supports both token-based and basic authentication modes.

• Use AIRFLOW_BASE_URL to point to your Airflow REST API host.
• If you use a token, set AIRFLOW_TOKEN and leave the username/password unset.
• If you use basic auth, set AIRFLOW_USERNAME and AIRFLOW_PASSWORD and omit the token.

Usage examples for day-to-day tasks

List all DAGs and pick one to inspect or manage.

Trigger a DAG run with optional configuration, pause or unpause a DAG, and fetch details about specific DAG runs or task instances.

Security and best practices

Store credentials securely and use environment variables or a secrets manager. Prefer API tokens for production deployments and enable TLS on your Airflow instance. Apply rate limiting and monitor usage to protect your Airflow environment.

Available tools

airflow_list_dags

List all DAGs with pagination and sorting

airflow_get_dag

Get detailed information about a specific DAG

airflow_trigger_dag

Trigger a new DAG run with optional configuration

airflow_pause_dag

Pause a DAG

airflow_unpause_dag

Unpause a DAG

airflow_list_dag_runs

List DAG runs for a specific DAG

airflow_get_dag_run

Get details of a specific DAG run

airflow_list_task_instances

List task instances for a DAG run

airflow_get_task_instance

Get detailed task instance information

airflow_get_task_logs

Get complete logs for a specific task instance

airflow_get_dag_run_logs

Get logs for all tasks in a DAG run

airflow_tail_dag_run

Tail/monitor a DAG run with recent activity and logs