A SQL Workspace
that is fully local.

Explore, visualize and interact with data from the browser or embedded in DuckDB as a community extension.

Visual exploration

Create interactive visualizations from your data with a simple SQL query.

Table, chart, text, dropdown, ...
Resize and reposition freely on the canvas

Chain Queries Together

Reference another node's output using dash.ref_<name>() in your SQL. Drag an edge between nodes and Dash inserts the reference automatically.

Downstream nodes re-run automatically
Dash figures out the correct execution order

Interactive Views

Each node forwards only its selected rows to the next node in the chain, letting you build fully interactive canvases.

Use a Drop Down to select rows
Instant re-execution on selection change

import duckdb

con = duckdb.connect()

con.execute("INSTALL dash FROM community; LOAD dash;")

con.execute("PRAGMA DASH;")

df = con.execute("SELECT * FROM dash.ref_monthly_revenue()").df()

Query Canvas Nodes

The same dash.ref_<name>() references work from any DuckDB connection, e.g. the CLI or directly from Python.

You can use Dash in two ways, either directly in your browser or as a DuckDB extension.

In Your Browser

Use Dash without any installation. All processing happens locally in your browser, so your data never leaves your device. The data is stored in your browser's file system.

As DuckDB Extension

Dash can be hosted within your DuckDB instance, allowing you to access local files and databases.

-- Install and load

INSTALL dash FROM community;

LOAD dash;

-- Start Dash

PRAGMA dash;