Skip to content

Welcome to dbt-score

dbt-score is a linter for dbt metadata.

dbt allows data practitioners to organize their data into models and sources. Those models and sources have metadata associated with them: documentation, tests, types, etc. dbt supports more types of entities, e.g. snapshots, analysis, seeds and more.

dbt-score allows to lint and score this metadata, in order to enforce (or encourage) good practices. The dbt entities that dbt-score is able to lint (currently) are:

  • Models
  • Sources
  • Snapshots

Example

> dbt-score lint --show all
🥇 M: customers (score: 10.0)
    OK   dbt_score.rules.generic.has_description
    OK   dbt_score.rules.generic.has_owner
    OK   dbt_score.rules.generic.sql_has_reasonable_number_of_lines
Score: 10.0 🥇

In this example, the model customers scores the maximum value of 10.0 as it passes all the rules. It also is awarded a golden medal because of the perfect score. By default a passing dbt entity with or without rule violations will not be shown, unless we pass the --show-all flag.

Philosophy

dbt entities are often used as metadata containers: either in YAML files or through the use of {{ config() }} blocks, they are associated with a lot of information. At scale, it becomes tedious to enforce good practices in large data teams dealing with many dbt entities.

To that end, dbt-score has 2 main features:

  • It runs rules on dbt entities, and displays any rule violations. These can be used in interactive environments or in CI.
  • Using those run results, it scores items, to ascribe them a measure of their maturity. This score can help gamify metadata improvements/coverage, and be reflected in data catalogs.

dbt-score aims to:

  • Provide a predefined set of good practices (the core rules).
  • Allow teams to easily add their own rules.
  • Allow rule sets to be packaged and distributed.
  • Be configurable to adapt to different data stacks and practices.

About

dbt-score is free software, released under the MIT license. It originated at Picnic Technologies in Amsterdam, Netherlands. Source code is available on Github.

All contributions, in the form of bug reports, pull requests, feedback or discussion are welcome. See the contribution guide for more information.