SQL Lineage Visualizer
✓
2
3
4
Dialect:
example.sql
Loading...
Simulation Logs
No active simulation session. Click "Run Simulation" above.
No lineage graph yet
Write SQL on the left, then click Parse DAG
Tables0
No tables yet.
Parse DAG or upload files.
Select a table or click a DAG node
Select a table from the list or click a node in the DAG

Understand any SQL pipeline at a glance

Paste stored procedures, ETL scripts, or view definitions and get an interactive lineage graph showing how data flows from source tables through staging steps into final outputs. The bundled example on first load shows the idea: two raw tables feed a staging table, which feeds a report table. Replace it with your own SQL and click Parse DAG. Then use Run Simulation to watch mock data flow through every step — all locally, without connecting to a database.

Frequently Asked Questions

How do I visualize my SQL lineage?
1) Paste your SQL into the editor (or add more tabs with the + button — one per file). 2) Pick the SQL dialect your scripts are written in. 3) Click "Parse DAG" to build the lineage graph. 4) Explore: drag to pan, scroll to zoom, and click any node to see its columns and details in the inspector panel.
What do the node colors in the graph mean?
Blue nodes are source tables your SQL reads from, orange nodes are temporary tables and views created along the way, purple nodes are CTEs (WITH clauses), and green nodes are final output tables written by INSERT, CREATE TABLE AS, or MERGE statements.
What does "Run Simulation" do?
It executes your statements in order against editable mock data, entirely in your browser via DuckDB WASM. You can see row counts flow through each step, click a node to preview its data, and edit the mock rows (or upload CSV/Parquet samples) in the Tables panel to test with realistic values. Nothing touches a real database.
Can I visualize multiple SQL files together?
Yes. Add a tab per file or upload several .sql files. Lineage is resolved across files — for example, a view defined in one file and referenced by a procedure in another shows up as a single connected graph.
Which SQL dialects are supported?
Snowflake, BigQuery, PostgreSQL, MySQL, Spark SQL, DuckDB, and generic ANSI SQL. Pick the dialect matching your scripts so dialect-specific syntax parses correctly.
Is my SQL sent anywhere?
Parsing happens on the bundled local backend on your machine, and simulation runs inside your browser. Your SQL and data never leave your computer.

Related Tools

SQL Query ToolRun SQL against local CSV/Parquet files.SQL FormatterBeautify and format SQL queries.DBT Lineage ViewerVisualize dbt project dependencies.