Astronomer

Translates "who signed up last month" or "which customers use feature X" into a warehouse query, runs it, and returns the results in context.

Generates an adapter layer that detects which Airflow version is running and routes API calls to the right method—no if-statements scattered through your codebase.

Queries DAGs, reads task logs, triggers runs, diagnoses failures, and checks connections via the airflow CLI. Surfaces the root cause of stuck tasks and failed DAGs in seconds.

Guides you through task definition, dependency graphs, and idempotency patterns—shows the common pitfalls that break a DAG in production and how to avoid them.

Generates an OpenLineage extractor for Airflow operators that don't ship one built-in, including column-level lineage and complex input/output mappings.

Converts a description of your data pipeline into a dag-factory YAML config that Airflow parses—no Python coding required. Handles task sequencing, defaults, and dynamic task generation.

Delegates Airflow, dbt, and data-pipeline questions to Otto, Astronomer's specialized agent. Handles migrations, upgrades, and debugging without losing context from your main task.

Takes a DAG or project directory, handles CI/CD setup, and deploys to your production Airflow instance with rollback steps and health checks built in.

Queries a table and generates a report showing nulls, cardinality, data types, sample values, and anomalies — so you know what you're working with before you query.

Maps every downstream table, DAG, and report that depends on a given data source, then flags which ones will break or misbehave if you modify it.

Pauses a data pipeline and waits for a human approval, form input, or manual decision before proceeding. Works with Airflow 3.1+ and handles branching based on the human choice.

Adds inlets and outlets to Airflow tasks so you can see which datasets feed which steps. Surfaces hidden dependencies and data quality gaps across your DAGs.

Inspects a table, dataset, or dashboard and reports the last update date, refresh cadence, and whether rows have aged beyond usefulness for your current decision.

Configures Astronomer Cosmos 1.11+ to run dbt Fusion projects locally against Snowflake or Databricks, with the right ExecutionMode and dependency checks baked in.

Spins up or tears down a local Airflow instance via Astro CLI, tails logs, and runs queries against the API to diagnose stuck DAGs or config problems.

Scaffolds a new Astro/Airflow project with dependencies wired, connections configured, and variables set. Hands you a ready-to-edit folder structure.

Analyzes logs, task sequences, and dependency chains to find the exact failure point in a DAG, then suggests the minimal fix and how to catch it next time.

Traces a table or metric back to its sources—showing every pipeline step, upstream dependency, and original feed. Surfaces breaks and missing links.

Scaffolds an Airflow 3.1+ plugin with FastAPI routes, custom UI pages, React components, and operator links. Handles boilerplate so you skip to the logic.

Scans your warehouse schema and generates a single reference doc with every table, column, type, and relationship. Query it instantly without running schema explorers.

Converts a simple YAML file into a validated, reusable Airflow DAG with typed task groups. Non-engineers can compose pipelines by editing config instead of writing Python.

Scans your Airflow project, replaces the old airflow-ai-sdk with apache-airflow-providers-common-ai, and updates all LLM decorators and imports to match the new API.

Converts a dbt Core project into an Airflow DAG or TaskGroup via Astronomer Cosmos, handling the warehouse, execution environment, and manifest setup with minimal manual rewiring.

Scans your Airflow 2 DAGs, flags breaking changes, rewrites deprecated operators and syntax, and generates a step-by-step upgrade plan with test checkpoints.

Executes a DAG, captures failures, diagnoses the root cause (bad config, missing dependency, logic error), and patches the code so it runs clean.