Connect your DuckDB

What you'll do

Open the .bruin.yml that dac init generated, repoint the local_duckdb connection at your existing DuckDB file, and verify the connection with dac connections and dac query.

Why this matters

DAC reuses Bruin CLI connections - the .bruin.yml you edit here is the same file the rest of the Bruin toolchain reads. Once a connection is wired in, every dashboard, query, and validation step picks it up automatically.

Instructions

1. Look at the connection block

cat .bruin.yml

The starter ships with a placeholder DuckDB connection:

default_environment: default

environments:
  default:
    connections:
      duckdb:
        - name: local_duckdb
          path: data/dac-demo.duckdb
          read_only: true

2. Point it at your DuckDB

Edit path: to wherever your DuckDB file lives. An absolute path works; a path relative to the project root works too. Leave read_only: true for dashboards - DAC only needs to SELECT.

duckdb:
  - name: local_duckdb
    path: /Users/you/data/warehouse.duckdb   # <- your file
    read_only: true

3. Verify the connection

dac connections --dir .

You should see:

NAME           TYPE     STATUS
local_duckdb   duckdb   ✓ connected

4. Run an ad-hoc query

Pick any table in your database. The example below assumes an orders table with a region and amount column - swap in your own:

dac query "SELECT region, SUM(amount) AS revenue FROM orders GROUP BY 1 ORDER BY 2 DESC" \
  --connection local_duckdb --dir .

DAC prints the result as a clean table. Use --output json or --output csv when you want to pipe it elsewhere.

What just happened

You proved the connection works end to end - from .bruin.yml through dac query to real rows in DuckDB. With a live connection in place, the next step is to put a dashboard on top of it.