Language: 日本語English

data Plugin

Category
Development
Topics
Data Engineering & Analytics · Databases & Storage · Migration & Modernization
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.

This item is "agents" by Astronomer, an Apache 2.0-licensed toolkit of AI agent tooling for data engineering workflows built around Apache Airflow. It bundles an MCP server (astro-airflow-mcp) for Airflow REST API integration, a terminal CLI tool called af, and a set of skills that extend AI coding agents (Claude Code, Cursor, and other agentic tools) with capabilities for authoring, testing, debugging, and deploying Airflow DAGs, tracing data lineage, profiling warehouse tables, integrating dbt via Astronomer Cosmos, and migrating from Airflow 2.x to 3.x. It works with open-source Apache Airflow as well as Astronomer's managed Astro platform. Installation is via npx skills add astronomer/agents for skills, or via the Claude Code plugin marketplace for the bundled astronomer-data plugin. Data warehouse connections support Snowflake, PostgreSQL, BigQuery, and 25+ other databases via SQLAlchemy.

Overview

"agents" is an open-source (Apache 2.0) repository maintained by Astronomer providing AI agent tooling for data engineering: an MCP server for the Airflow REST API, a CLI (af), and a library of "skills" that plug into agentic coding tools such as Claude Code and Cursor. It is not itself a hosted SaaS; it is installed locally or as a Claude Code plugin, and can connect to either self-hosted Apache Airflow or Astronomer's managed Astro platform.

What you can do with data

  • Install the MCP server (astro-airflow-mcp) to manage Airflow DAGs, trigger runs, view task logs, and check system health via the Airflow REST API
  • Use the af CLI for terminal commands like af health, af dags list, or af runs trigger
  • Initialize warehouse schema discovery (warehouse-init) and run SQL-based analysis (analyzing-data) against connected data warehouses (Snowflake, PostgreSQL, BigQuery, DuckDB, Redshift, and other SQLAlchemy-supported databases)
  • Check data freshness (checking-freshness) and run comprehensive table profiling and quality assessment (profiling-tables)
  • Trace upstream/downstream data lineage, annotate task lineage with inlets/outlets, and build custom OpenLineage extractors
  • Set up, author, test, and debug Airflow DAGs, including composing DAGs from YAML via blueprint and running dbt Core/Fusion projects as DAGs via Astronomer Cosmos
  • Deploy Airflow DAGs and projects to Astro, Docker Compose, or Kubernetes, and use human-in-the-loop workflows (airflow-hitl) for Airflow 3.1+
  • Migrate DAGs from Airflow 2.x to Airflow 3.x using the migrating-airflow-2-to-3 skill

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 data

Back to list