Astronomer's toolkit for Apache Airflow and the data it moves. Ask for a pipeline, a fix or an answer from your warehouse, and the plugin follows the steps an Airflow engineer would.
Build and test DAGs. Write new DAGs, extend existing ones, and test them locally before you deploy. Covers dbt with Cosmos, DAG Factory, Blueprint, human-in-the-loop tasks, Airflow plugins, and tasks written in Go or Java.
Debug failures. Find out why a DAG fails to parse or a task keeps failing. The plugin reads the logs, finds the root cause, and fixes the code.
Run Astro. Set up an Astro project, run Airflow locally, deploy, and troubleshoot Astro deployments.
Migrate. Move DAGs from Airflow 2 to Airflow 3, or move a Dagster project to Airflow.
Query your warehouse. Answer business questions with SQL against Snowflake, BigQuery, Postgres and other SQLAlchemy databases. Profile tables, check how fresh the data is, and generate a schema reference so later questions go faster.
Trace lineage. See what feeds a table and what breaks downstream if it changes.
Most skills run local tools such as the Astro CLI and uv, so they work best in Codex. Warehouse skills read connection details from ~/.astro/agents/warehouse.yml.