Apache Airflow MCP Server Integration

What does “Apache Airflow” MCP Server do?

The Apache Airflow MCP Server is a Model Context Protocol (MCP) server that acts as a bridge between AI assistants and Apache Airflow instances. By wrapping Apache Airflow’s REST API, it enables MCP clients and AI agents to interact with Airflow in a standardized and programmatic way. Through this server, developers can manage Airflow DAGs (Directed Acyclic Graphs), monitor workflows, trigger runs, and perform various workflow automation tasks. This integration streamlines development workflows by allowing AI-driven tools to query the state of data pipelines, orchestrate jobs, and modify workflow configurations directly via MCP. The server leverages the official Apache Airflow client library to maintain compatibility and ensure robust interaction between AI ecosystems and Airflow-powered data infrastructure.

List of Prompts

No explicit prompt templates are documented in the available files or repository content.

List of Resources

No explicit MCP resources are documented in the repository content or README.

List of Tools

• List DAGs
  Allows clients to retrieve a list of all DAGs (workflows) managed by the Airflow instance.
• Get DAG Details
  Retrieve detailed information about a specific DAG identified by its ID.
• Pause DAG
  Pauses a specific DAG, preventing scheduled runs until unpaused.
• Unpause DAG
  Unpauses a specific DAG, allowing it to resume scheduled execution.
• Update DAG
  Update configuration or properties of a specific DAG.
• Delete DAG
  Remove a specific DAG from the Airflow instance.
• Get DAG Source
  Fetch the source code or file contents of a given DAG.
• Patch Multiple DAGs
  Apply updates to multiple DAGs in a single operation.
• Reparse DAG File
  Trigger Airflow to reparse a DAG file, useful after code changes.
• List DAG Runs
  List all runs for a specific DAG.
• Create DAG Run
  Trigger a new run for a specific DAG.
• Get DAG Run Details
  Get detailed information about a particular DAG run.

Use Cases of this MCP Server

• Automated Workflow Orchestration
  Developers can use AI agents to schedule, trigger, and monitor Airflow workflows programmatically, reducing manual intervention and increasing automation.
• DAG Management and Version Control
  AI assistants can help manage, pause, unpause, and update DAGs, making it easier to handle complex pipeline lifecycles and changes.
• Pipeline Monitoring and Alerting
  The server allows AI tools to query the state of DAG runs, enabling proactive monitoring and alerting on workflow failures or successes.
• Dynamic DAG Modification
  Enables dynamic updates or patching of DAGs based on real-time requirements, such as changing schedules or parameters.
• Source Code Inspection and Debugging
  AI tools can retrieve DAG source files for code review, debugging, or compliance checks directly from the Airflow instance.

How to set it up

Windsurf

1. Ensure you have Node.js and Windsurf installed on your machine.
2. Locate the Windsurf configuration file (commonly windsurf.config.json).
3. Add the Apache Airflow MCP Server to the mcpServers section:

   {
     "mcpServers": {
       "apache-airflow": {
         "command": "npx",
         "args": ["@yangkyeongmo/mcp-server-apache-airflow@latest"]
       }
     }
   }
   

4. Save the configuration file.
5. Restart Windsurf and verify the Airflow MCP Server loads successfully.

Securing API Keys Example:

{
  "mcpServers": {
    "apache-airflow": {
      "command": "npx",
      "args": ["@yangkyeongmo/mcp-server-apache-airflow@latest"],
      "env": {
        "AIRFLOW_API_KEY": "your-airflow-key"
      },
      "inputs": {
        "api_url": "https://your-airflow-instance/api/v1/"
      }
    }
  }
}

Claude

1. Ensure Node.js is installed and Claude’s config file is accessible.
2. Edit the configuration file to include the Apache Airflow MCP Server.
3. Use the following JSON snippet:

   {
     "mcpServers": {
       "apache-airflow": {
         "command": "npx",
         "args": ["@yangkyeongmo/mcp-server-apache-airflow@latest"]
       }
     }
   }
   

4. Save and restart Claude.
5. Confirm connection and functionality.

Cursor

1. Verify Node.js installation.
2. Open Cursor’s configuration file.
3. Add:

   {
     "mcpServers": {
       "apache-airflow": {
         "command": "npx",
         "args": ["@yangkyeongmo/mcp-server-apache-airflow@latest"]
       }
     }
   }
   

4. Save and restart Cursor.
5. Check the MCP Server integration.

Cline

1. Install Node.js if not present.
2. Navigate to Cline’s configuration file.
3. Insert:

   {
     "mcpServers": {
       "apache-airflow": {
         "command": "npx",
         "args": ["@yangkyeongmo/mcp-server-apache-airflow@latest"]
       }
     }
   }
   

4. Save and restart Cline.
5. Verify the MCP Server connection.

Note: Secure your Airflow API keys using environment variables as shown in the Windsurf example above.

How to use this MCP inside flows

Using MCP in FlowHunt

To integrate MCP servers into your FlowHunt workflow, start by adding the MCP component to your flow and connecting it to your AI agent:

Click on the MCP component to open the configuration panel. In the system MCP configuration section, insert your MCP server details using this JSON format:

{
  "apache-airflow": {
    "transport": "streamable_http",
    "url": "https://yourmcpserver.example/pathtothemcp/url"
  }
}

Once configured, the AI agent is now able to use this MCP as a tool with access to all its functions and capabilities. Remember to change “apache-airflow” to the actual name of your MCP server and replace the URL with your server’s URL.

Overview

Our opinion

The Apache Airflow MCP Server provides robust tooling for workflow management and automation, but lacks documentation on prompt templates and explicit MCP resources. Its setup is straightforward, and the presence of a MIT license and active development are positives. However, lack of sampling and roots feature documentation slightly limits its scope for agentic LLM workflows.

MCP Score