---
id: dbt-colibri
version: "0.3.6"
license: MIT
license_treatment: permissive
maintenance: active
---

# dbt-colibri — A column lineage parser and dashboarding tool

License: permissive · Maintenance: active · Downloads: 90.3K/mo

## What it is and what it does

dbt-colibri is a lightweight CLI tool that parses dbt project artifacts to extract column-level lineage—the path each column takes through your dbt transformations—and renders it as an interactive, self-contained HTML dashboard. It uses SQLGlot to parse SQL and extract column dependencies, then combines that with dbt's manifest and catalog metadata to build a complete lineage graph. The tool generates a static HTML file that requires no external services, cloud dependencies, or complex setup.

You run it after your normal dbt compile and dbt docs generate steps, point it at your target directory, and get back a browsable visualization of how data flows column-by-column through your models. It's designed for data teams who need transparency into their transformation logic without vendor lock-in or enterprise tooling overhead. The package supports a wide range of dbt adapters (Snowflake, BigQuery, Redshift, duckDB, Postgres, Databricks, Athena, Trino, SQL Server, ClickHouse, Oracle, StarRocks) and has been tested against dbt-core 1.8.x through 1.10.x.

Use it for:

- Generate column lineage reports for dbt projects to understand data dependencies before refactoring models.
- Deploy lineage dashboards to GitHub Pages or internal web servers via CI/CD for team visibility.
- Audit data quality by tracing which source columns feed into critical downstream tables.
- Document dbt project structure for onboarding new team members without manual diagram creation.
- Identify unused columns or dead-end transformations by visualizing the complete column flow.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Extracts column-level lineage from dbt projects and generates an interactive HTML dashboard showing how data flows through your transformations.

Yes. Low install friction, no security vulnerabilities, permissive MIT license, and active maintenance make this a straightforward choice for dbt users needing column-level lineage visibility. The self-hosted, static-HTML approach eliminates vendor lock-in concerns. Install if your team uses dbt and wants transparent, flexible lineage tracking.

## Install

pip install dbt-colibri
uv add dbt-colibri
poetry add dbt-colibri

## Installing dbt-colibri

Before you install: Low friction install with only two runtime dependencies (click and sqlglot). Actively maintained with a recent release, supporting Python 3.9 through 3.13.

License in practice: MIT license permits unrestricted use, modification, and distribution with minimal obligations—suitable for commercial and private projects.

Quickstart:

pip install dbt-colibri

# After running: dbt compile && dbt docs generate
colibri generate
# Opens dist/index.html in browser

Requires dbt artifacts (manifest.json and catalog.json) generated by running dbt compile and dbt docs generate first.

Verify before relying:

- Whether the tool handles all dbt model types equally or has limitations beyond the noted Databricks SQL-only constraint.
- Performance characteristics and HTML file size for projects with thousands of models or columns.

## Package facts

- License: MIT (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 90.3K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

dbt column lineage visualization, dbt data lineage tracking, column-level lineage extraction, dbt dashboard self-hosted, dbt metadata analysis tool, data lineage reporting, dbt artifact visualization, dbt-ecosystem, data-lineage, self-hosted

[View on SkillFed](https://skillfed.io/packages/dbt-colibri) · [View on PyPI](https://pypi.org/project/dbt-colibri/)