How to Visualize a dbt Project Without dbt Docs

See your dbt lineage graph without dbt docs generate or a warehouse connection. Drop your models folder and view it instantly. Free, nothing uploaded.

Quick answer: Free dbt project visualizer and SQL file lineage tool.

Related: SQL Lineage Analyzer for a Folder of SQL Files · dbt Lineage Viewer for Your SQL Models · SQL Column Lineage Tool · SQL File Dependency Analyzer · dbt Project Visualizer and dbt Lineage Guides

How to use the How to Visualize a dbt Project Without dbt Docs

Free dbt project visualizer and SQL file lineage tool.

The SQL Project Lineage Analyzer is a free SQL file lineage tool and SQL query visualizer. It reads a whole folder of SQL files, a dbt project or a ZIP, and draws how the tables connect. For every CREATE TABLE, CREATE VIEW, CREATE TABLE AS, INSERT … SELECT, MERGE, UPDATE … FROM and SELECT INTO it finds the tables read and the table written, follows CTEs, subqueries, joins and UNIONs, and traces each output column back to the source columns it was built from, across files. dbt ref() and source() calls are followed too, so a models folder works without running dbt. You get a graph in the same style as the Data Lineage tool, a table list with how many tables each one feeds, column lineage you can trace upstream and downstream, a suggested run order for the files, and a list of problems such as circular dependencies. Save the graph as PNG or JPG, copy it, or share a link. Everything happens in your browser; your SQL is not uploaded.

  1. Drop a folder, a ZIP or a GitHub link
  2. Click a table to see what feeds it and what it affects
  3. Open Columns to trace one column

Tips

Example

Input: a dbt models folder, no manifest.json

Output: The full lineage graph, drawn from the SQL files alone

Browse by topic

Who uses Data Utility

Frequently asked questions · Data format benchmark

Frequently asked questions

Can I see dbt lineage without dbt docs generate?

Yes. This tool reads the model SQL directly and follows ref() and source(), so no manifest.json and no warehouse connection are needed.

What does dbt docs show that this does not?

dbt docs also shows descriptions, tests and compiled SQL from your YAML files and a live warehouse. This tool reads .sql files only, so it focuses on the graph and column links.

How do I use this SQL lineage tool?

Drop a folder of SQL files, a ZIP or a dbt project onto the page, or paste a query. The graph appears at once. Click any table to see what it is built from and what depends on it, then open Columns to follow a single column.

Can I visualize my dbt project with it?

Yes. Drop your dbt models folder (or a ZIP of the project). ref() and source() are followed, so sources, staging models and marts appear as a graph without running dbt or connecting a warehouse. It reads .sql files for the lineage, and your .yml schema files for descriptions and tests, which then show on each table and column. Macro files and project settings are skipped.

Does it read my dbt YAML files?

Yes. Drop the whole dbt project folder or ZIP. The .yml schema files (models, sources, seeds, snapshots) are read for descriptions and tests. They show when you click a table, in the Columns tab and in the Tables tab. dbt_project.yml, packages.yml and macro files are not documentation and are skipped. A schema file that cannot be parsed is listed under the stats.

Can I edit my SQL and see what the change would break?

Yes. Under the graph, open Project files, click a file and press Edit. The graph updates as you type, and a card called What your edits would affect lists what would break downstream: removed or renamed columns that something still reads, changed data types and calculations, removed tests, and personal data reaching new columns. Copy summary gives you Markdown for a pull request. Undo all edits restores the originals, and your own files are never changed.

Can I follow personal data (PII) through my SQL?

Yes. Columns with names like email, phone or address are tagged automatically, and you can tag your own with Tag as PII in the Columns tab. The PII tab then lists every table that holds data built from them, and marks columns that are hashed, masked or only counted as protected. It is a guess from column names, so check the tags.

Can I load a project straight from GitHub?

Yes, for public repositories. Paste the repository link (or a link to a branch, tag or folder, such as github.com/owner/repo/tree/main/models) and click Load from GitHub. Your browser lists the files and downloads the .sql and .yml files directly from GitHub, then builds the lineage. Private repositories are not available because no sign-in is used, and GitHub limits anonymous requests, so after many imports in an hour you may need to wait or drop a ZIP instead.

Is my SQL uploaded?

No. The files you drop are read and analysed in your browser and are not sent anywhere. If you load a public GitHub repository, your browser downloads the files from GitHub itself; they still do not pass through our servers.

Can I visualize a single SQL query?

Yes. Paste the query into the box. You see the tables it reads and the table it writes, and for a CREATE VIEW or INSERT … SELECT the column links too.

How accurate is the column lineage?

It is read from the SQL, following aliases, casts, CASE, functions and subqueries. It is best effort. Dynamic SQL built from strings, vendor-only syntax and macros that generate table names cannot be followed.

Which SQL dialects work?

The common ANSI core used by Postgres, Snowflake, BigQuery, Redshift, Databricks, MySQL and SQL Server, including quoted names in "double quotes", backticks and [brackets]. Unreadable statements are listed instead of being skipped silently.

How do I save or share the graph?

Open the Export tab. Save a PNG or JPG, copy the picture to paste into a document, or copy a share link that opens the same lineage for anyone you send it to.

How can I see lineage for a folder of SQL files for free?

Open SQL Project Lineage and drop the folder, a ZIP or a dbt models folder. The table and column lineage is drawn in your browser, with a run order and an impact view. It is free, needs no sign-up and nothing is uploaded.

Is there a free dbt project visualizer?

Yes. Drop your dbt models folder into SQL Project Lineage. It follows ref() and source() and draws sources, staging models and marts as a clickable graph with column lineage. You do not need to run dbt, and nothing is uploaded.

Is there a free SQL query visualizer?

Yes. Paste a query or drop SQL files into SQL Project Lineage and it draws the tables, joins and column links as a diagram. You can save it as PNG or JPG, copy it, or share a link.

Related pages

Topic guides

Guides