Airflow MCP Server

MCPOpen SourceMIT10.0

by treerootboy • Automation & Orchestration

An MCP server that integrates with Apache Airflow to trigger, monitor, and manage DAG runs through the Airflow REST API.

Example Use Cases

1
Trigger Airflow DAG runs programmatically and pass configuration to jobs.
2
Collect DAG run status, list DAGs and DAG runs, and aggregate daily reports for monitoring or auditing.
3
Fetch task logs or perform backfills across date ranges to recover or reprocess data.

Description

This project implements a Model Context Protocol (MCP) server for Apache Airflow, providing tools to trigger DAGs, enable DAGs, fetch DAG statuses, list DAGs and DAG runs, retrieve logs, and perform backfills. It integrates with the Airflow REST API and is intended to simplify automation and monitoring of Airflow workflows. The repository includes instructions for local development and running an Airflow instance via docker-compose. It is open-source under the MIT license.

Capabilities(9 total)

Tools (8)

trigger-dag

Trigger a DAG run

dag_id:string*conf:object

enable-dag

Enable a DAG

dag_id:string*

get-daily-report

Get a consolidated report for all DAG runs within specified time range

start_date:stringdays:integer

list-dags

List all available DAGs

list-dag-runs

Get DAG runs in batch

dag_ids:arrayexecution_date_gte:stringexecution_date_lte:stringstates:arraylimit:integeroffset:integerorder_by:string

Prompts (1)

summarize-dags

Creates a summary of all DAGs

style

Quick Actions

View on GitHub

Security

Scanned 2 month(s) ago

Risk Level

MINIMAL

Read-only data retrieval, no side effects

Trust Score

B67/100

10/17 checks passed

Scores are informational only and provided “as is” without warranty. AgentHotspot assumes no liability for actions taken based on these ratings.

Quick Stats

Service TypeMCP

Pricing ModelFree

Capabilities8 Tools / 1 Prompts / 0 Resources

Ownertreerootboy

CategoryAutomation & Orchestration

DependenciesStandalone

Set Your Username