openlineage-dbt

What it is and what it does

OpenLineage dbt is a wrapper that intercepts dbt runs and automatically captures metadata about model execution, inputs, and outputs, then sends that lineage data to an OpenLineage backend. It replaces the standard `dbt run` command with `dbt-ol run`, allowing you to track data flows and dependencies across your dbt project without modifying your dbt code. The package depends on dbt-core and the openlineage-integration-common library to handle transport and formatting.

It works with a wide range of dbt adapters (bigquery, snowflake, spark, redshift, athena, glue, postgres, clickhouse, trino, databricks, sqlserver, fabric, dremio, duckdb) and is configured via environment variables pointing to your OpenLineage backend. Optional table and column-level metadata is available if you've run `dbt docs generate` to produce a catalog.json file.

Use it for:

• Track data lineage across dbt models in a centralized OpenLineage instance for compliance and impact analysis.
• Automatically log model run lifecycle events without modifying dbt projects or adding custom hooks.
• Integrate dbt metadata into a broader data governance or observability platform that consumes OpenLineage events.
• Capture input/output relationships for dbt models running on cloud data warehouses (Snowflake, BigQuery, Redshift, Databricks).
• Send historical run metadata using `dbt-ol send-events` to backfill lineage without re-running jobs.

Worth the install?

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

Automatically collects and sends dbt run metadata—model lifecycle, inputs, and outputs—to an OpenLineage backend for data lineage tracking.

Yes, if you run dbt and need to track lineage in an OpenLineage backend. Low install friction, active maintenance, no known vulnerabilities, and a clear use case for data governance. The main gotcha is the unclear license metadata—verify the Apache-2.0 claim before use in restricted contexts—and ensure your dbt adapter is in the supported list.

Install

openlineage-dbt on PyPI

pip install openlineage-dbt

uv add openlineage-dbt

poetry add openlineage-dbt

Installing openlineage-dbt

Before you install

Low install friction with a pure-Python wheel. Active maintenance (released 22 days ago). Depends on tqdm, openlineage-integration-common, and dbt-core; the dbt-core dependency will pull in its own ecosystem.

License in practice

License treatment is unclear in the metadata; the description footer notes Apache-2.0 and copyright 2018-2026 to the OpenLineage project, but this is not formally declared in the package metadata. Verify the actual license before use in proprietary or restricted contexts.

Quickstart

pip install openlineage-dbt

# Set OpenLineage backend URL
export OPENLINEAGE_URL=http://localhost:5000

# Replace 'dbt run' with 'dbt-ol run'
dbt-ol run

Requires dbt >= 0.20 and Python >= 3.10 (per requires_python). Only supports specific dbt adapters: bigquery, snowflake, spark, redshift, athena, glue, postgres, clickhouse, trino, databricks, sqlserver, fabric, dremio, duckdb.

Verify before relying

• Whether the Apache-2.0 license stated in the description footer is the authoritative license, given the unclear metadata treatment.
• Whether catalog.json (from 'dbt docs generate') is required or optional for full metadata collection.
• Support status and compatibility with dbt versions newer than 0.20.

Package facts

Evidence: openlineage_dbt-1.52.0-py3-none-any.whl

Keywords: openlineage

Tags

More Database packages