Why Does this Exist

With dbt, data teams work directly within the warehouse to produce trusted datasets for reporting, ML modeling, and operational workflows. Part of this workflow includes managing YAML files and metadata. Little things like column descriptions, tags, and tests can be a pain to maintain and tend not to be DRY.

What Can we Automate

Most of the time, you don't need to write YAML by hand. dbt-osmosis automates the process of generating YAML files for you. You can focus on what matters. Source yaml files are generated from your database. Schema yaml files are generated from your dbt models with inherited metadata.

Git Oriented

A single execution on your dbt project can save you hours of manual toil. dbt-osmosis is built to work with your existing dbt project. It can be run directly, as a pre-commit hook, or as a CI/CD step generating a PR. By leveraging git, we can safely execute file changes en masse with proper diffs and version control.