dbt-score

What it is and what it does

dbt-score is a linting and scoring tool for dbt projects that evaluates model metadata quality against configurable rules. It assigns numerical scores (0-10) to individual models and computes an overall project score, helping data teams enforce consistent documentation, testing, naming, and structural standards across their dbt codebase.

The tool runs from the command line, reads your dbt manifest, and reports violations organized by severity and rule type. It integrates with CI/CD pipelines by exiting with success or failure status based on configurable thresholds, and supports selective linting via dbt's selection syntax. Configuration happens in pyproject.toml, where you can disable rules, adjust severity levels, customize scoring badges, and set pass/fail thresholds for both individual models and the overall project.

Use it for:

• Enforce documentation standards in CI/CD by failing builds when models lack descriptions or owners
• Track data quality improvements over time by monitoring project scores and individual model ratings
• Prevent technical debt by catching overly long SQL queries and models without adequate test coverage
• Create custom rules to enforce organization-specific naming conventions or governance metadata
• Lint only recently changed models or specific project sections using dbt selection syntax

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

dbt-score lints dbt models against configurable rules for documentation, testing, naming, and structure, assigning numerical quality scores to individual models and projects.

Yes. dbt-score addresses a real gap in dbt project governance—automated enforcement of metadata quality at scale. It has low install friction, permissive licensing, no known vulnerabilities, active maintenance, and integrates naturally into existing dbt workflows. Install it if your team maintains multiple dbt models and wants programmatic quality gates; skip it only if your project is too small to benefit from standardized linting.

Install

dbt-score on PyPI

pip install dbt-score

uv add dbt-score

poetry add dbt-score

Installing dbt-score

Before you install

Installs cleanly with only two lightweight runtime dependencies (click and tomli). Marked as active maintenance with a recent release in 2026.

License in practice

MIT license is permissive; you can use, modify, and distribute this package freely with minimal restrictions.

Quickstart

pip install dbt-score

# From your dbt project root:
dbt-score lint

# With configuration in pyproject.toml:
[tool.dbt-score]
fail_project_under = 7.5
fail_any_item_under = 8.0

Requires Python 3.10 or later and dbt-core 1.5 or later installed in the same environment.

Verify before relying

• Whether the built-in rule set covers your organization's specific data quality standards or if custom rules are required
• Performance characteristics when linting large dbt projects with hundreds or thousands of models
• Integration experience with specific dbt-core versions beyond the stated 1.5+ requirement

Package facts

Evidence: dbt_score-0.16.0-py3-none-any.whl

Tags

More Quality Assurance packages

Further reading