Quickstart¶
~10 minutes · ends with the dbt DAG imported, a certified block, a manifest, and end-to-end lineage
DQL adds an analytics layer on top of a dbt project: certified blocks, dashboards in Apps, notebooks, and governed agent answers — all as files in git. This quickstart adds DQL to your own dbt repo. No external warehouse, hosted account, SSO, or team RBAC is required.
Just want to see it running? The jaffle-shop-duckdb repo ships a complete, ready-built DQL workspace (10 certified blocks + an executive App dashboard). Clone it,
./setup.sh, thencd dql && npm install && npm run notebook. The rest of this page builds the same thing from scratch on your repo.
1. Make your dbt manifest fresh¶
cd your-dbt-repo
dbt parse # or dbt build — either writes target/manifest.json
2. Scaffold the DQL workspace¶
From the dbt repo root:
npx create-dql-app@latest dql
cd dql
npm install
The scaffolder detects the parent dbt project and wires it into
dql.config.json. DQL stays isolated under ./dql; your dbt files are
untouched. The generated project installs the DQL CLI locally, so npm run
... works without a global dql binary.
Point the default connection at your dbt warehouse in dql.config.json. For a
local DuckDB file that's:
"connections": {
"default": { "driver": "duckdb", "filepath": "../my_warehouse.duckdb" }
}
(Postgres, Snowflake, BigQuery, etc. — see connectors.)
3. Check the setup and sync dbt¶
npm run doctor
npm run sync
dql doctor verifies the project shape, notebook assets, and default
connection. dql sync dbt imports the dbt models and sources into the DQL
manifest so lineage connects end to end.
4. Start the notebook¶
npm run notebook
The CLI starts a local server on http://127.0.0.1:3474 and opens the browser UI. Query a dbt model from a notebook cell, then open Blocks to create your first block from a dbt model (the authoring tutorial walks through a full certified block).
5. Compile the manifest and view lineage¶
npm run compile
npm run lineage
dql compile writes dql-manifest.json, the dbt-like artifact for the DQL
workspace: blocks, notebooks, Apps, dashboard pages, semantic objects, dbt
imports, and lineage edges. npm run lineage summarizes how data flows from
sources through dbt models into blocks, dashboard pages, and Apps.
Verify it worked¶
You should have:
- A running local notebook at
http://127.0.0.1:3474 - The dbt DAG visible in the Lineage view
dql-manifest.jsonwritten in the workspace root
Where to go next¶
- Tutorials — end to end (blocks → dashboards & Apps → agent → CI)
- DQL in 5 concepts
- Block Studio dbt-first workflow
- Import your dbt project