cosmos-dbt-fusion

Configure Astronomer Cosmos to execute dbt Fusion projects on Airflow.

3|Updated Jul 22, 2025
One-click install
npx skills add https://github.com/minyeamer/linkmerce --skill cosmos-dbt-fusion-minyeamer
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cosmos-dbt-fusion
Source: https://github.com/minyeamer/linkmerce/tree/main/.agents/skills/cosmos-dbt-fusion
Command: npx skills add https://github.com/minyeamer/linkmerce --skill cosmos-dbt-fusion-minyeamer

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

It solves configuration mistakes when running dbt Fusion projects on Airflow using Astronomer Cosmos, especially around Cosmos version requirements, the need for the dbt Fusion binary, and Fusion-only constraints like LOCAL execution.

Core Features & Use Cases

  • Fusion-only guardrails: validates Fusion-specific limitations such as no async and no virtualenv usage, plus supported warehouses (Snowflake, Databricks, BigQuery, Redshift).
  • Binary + execution wiring: guides you to install the dbt Fusion binary into the Airflow runtime and point Cosmos to it via ExecutionConfig for LOCAL execution.
  • Correct rendering/loading strategy: helps you select a parsing strategy (automatic, dbt_manifest, or dbt_ls) aligned with Fusion behavior and project size/selection needs.
  • Use case: You have an existing Airflow deployment using Cosmos and want to run a dbt Fusion project against Snowflake on a schedule, with reliable parsing and operator configuration.

Quick Start

Follow the Cosmos + dbt Fusion checklist to configure ExecutionMode.LOCAL and set dbt_executable_path to your installed Fusion binary, then assemble a DbtDag or DbtTaskGroup that targets your warehouse connection.

Frequently Asked Questions about cosmos-dbt-fusion

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run a dbt Fusion project with Astronomer Cosmos on Airflow?

To run dbt Fusion with Cosmos on Airflow, configure ExecutionMode.LOCAL, install the dbt Fusion binary into the Airflow runtime, and point Cosmos to it via the ExecutionConfig dbt_executable_path. Assemble a DbtDag or DbtTaskGroup targeting your warehouse connection.

What are the limitations when using dbt Fusion with Cosmos?

dbt Fusion with Cosmos enforces LOCAL-only execution constraints and prohibits async or virtualenv usage. Supported warehouses include Snowflake, Databricks, BigQuery, and Redshift. These guardrails prevent configuration mistakes when wiring Fusion into Airflow DAGs.

Does dbt Fusion work with Cosmos virtualenv or async execution modes?

No, dbt Fusion does not support async or virtualenv execution modes in Cosmos. It is constrained to ExecutionMode.LOCAL, requiring the dbt Fusion binary to be installed directly in the Airflow runtime rather than isolated in a virtual environment.

What Cosmos version is needed to execute dbt Fusion projects?

Executing dbt Fusion projects requires Cosmos version 1.11 or higher. This version provides the necessary ExecutionConfig and ProfileConfig rendering support to correctly wire the external dbt Fusion binary for LOCAL execution on Airflow.

How do I choose a parsing strategy for dbt Fusion in Airflow DAGs?

Select a parsing strategy between automatic, dbt_manifest, or dbt_ls based on your project size and selection needs. The choice must align with dbt Fusion behavior to ensure reliable DAG or TaskGroup generation within your Airflow deployment.

Can I use dbt Fusion with Snowflake on a scheduled Airflow DAG?

Yes, Snowflake is a supported warehouse for dbt Fusion execution through Cosmos. You can schedule a DbtDag or DbtTaskGroup by configuring a Fusion-compatible ProfileConfig that renders your Snowflake connection for LOCAL execution.