AI agent tooling for data engineering workflows.
This repository provides AI agent tooling for data engineering workflows, specifically an MCP (Model Context Protocol) server that enables AI assistants to interact with Apache Airflow. The project is built in Python and consists of two main components: a Claude Code plugin for workflow automation and an Airflow MCP server that supports both Airflow 2.x and 3.x through an adapter pattern. The tech stack includes FastMCP, httpx, and Pydantic, and it's designed to work with AI assistants like Claude, Cursor, and other MCP clients. Astronomer is the organization behind this project, as evidenced by the repository namespace and configuration references.