Language: 日本語English

astronomer-data-agents Plugin

Category
Development
Topics
Data Engineering & Analytics · Migration & Modernization · AI Agents & AI App Development
First seen
2026-07-09
Last confirmed
2026-07-13
Explanation last updated
2026-07-12

The explanation below is AI-generated. Please verify it against the sources.

astronomer-data-agents (GitHub: astronomer/agents) is an open-source, Apache 2.0-licensed toolkit built by Astronomer that provides AI agent tooling for Apache Airflow data engineering workflows. According to the README, it bundles an MCP server (astro-airflow-mcp) for Airflow REST API integration, a CLI tool called af for terminal-based Airflow interaction, and a set of skills that extend AI coding agents such as Claude Code and Cursor. These skills cover DAG authoring, testing, and debugging, data lineage tracing, table profiling, dbt/Cosmos integration, and Airflow 2-to-3 migration. The README states it works with self-hosted open-source Airflow (2.x/3.x REST API) as well as Astronomer's managed Astro platform, and skills are tool-agnostic across 25+ AI coding agents.

Overview

This is not a SaaS product itself but an open-source GitHub repository distributed as a Claude Code plugin, Cursor/other-agent skills, and an MCP server; it adds AI agent capabilities on top of Apache Airflow and optionally Astronomer's managed Astro platform, per the README.

What you can do with astronomer-data-agents

  • Install an MCP server (astro-airflow-mcp) giving AI agents full Airflow REST API access: DAG management, triggering, task logs, and system health
  • Use the af CLI to interact with Airflow from the terminal (e.g., af health, af dags list, af runs trigger )
  • Run data discovery/analysis skills: warehouse-init (schema discovery), analyzing-data (SQL-based Q&A via background Jupyter kernel), checking-freshness, profiling-tables
  • Trace data lineage: tracing-downstream-lineage, tracing-upstream-lineage, annotating-task-lineage, creating-openlineage-extractors
  • Author, test, and debug Airflow DAGs with best practices, and deploy them via Astro, Docker Compose, or Kubernetes (setting-up-astro-project, managing-astro-local-env, authoring-dags, blueprint, testing-dags, debugging-dags, deploying-airflow, airflow-hitl)
  • Run dbt Core or dbt Fusion projects as Airflow DAGs using Astronomer Cosmos (cosmos-dbt-core, cosmos-dbt-fusion)
  • Migrate DAGs from Airflow 2.x to 3.x using the migrating-airflow-2-to-3 skill
  • Connect to data warehouses (Snowflake, PostgreSQL, BigQuery, and 25+ databases via SQLAlchemy) through configuration files for schema discovery and querying

Sources

Original description (English)

Data engineering for Apache Airflow and Astronomer. Author DAGs with best practices, debug pipeline failures, trace data lineage, profile tables, migrate Airflow 2 to 3, and manage local and cloud deployments.

History of astronomer-data-agents

Back to list