Databao docs

Quickstart

Before you begin

Make sure you have the following set up:

  1. Python 3.11 or later
  2. uv installed
  3. Claude Code installed and configured
  4. A Google account
  5. A directory containing your dbt project

Run Databao CLI

  1. In the terminal, navigate to the root of your dbt project.

  2. Run Databao

    Terminal
    uvx databao

    This command runs the latest version of Databao CLI without installing it. To install Databao on your PATH, run uv tool install databao instead.

Sign in

  1. In the browser tab that opens, click Sign in with Google and follow the prompts.

  2. When prompted, return to the terminal.

Generate semantic layer

  1. In the terminal, select Start Databao in Claude Code:

    Terminal
    ? What would you like to do? (Use arrow keys)
» Start Databao in Claude Code
Advanced settings
Quit

  2. When asked by Claude, agree to use the skill and follow the prompts to generate the semantic layer.

    Claude may also ask you if you want to proceed with running certain commands.

    Generating the semantic layer may take some time depending on the complexity of your project.

Review and commit the semantic layer

  1. When the semantic layer is generated, Claude provides a summary of the work done:

    Output example
    ⏺ Here's the cycle summary:

Cycle summary
─────────────
Intent:         "bootstrap semantic layer with 5 metrics"
Files added:    7
models/semantic/time_spine_daily.sql
models/semantic/time_spine_daily.yml
models/semantic/sem_orders.yml
models/semantic/sem_customers.yml
models/semantic/fct_payments.sql
models/semantic/sem_payments.yml
Files modified: 1
.databao/test_questions.csv
Test set:       5 questions covered (was 0)
Validation:     dbt parse ✓, mf validate-configs ✓

Questions added:

1. What is the total revenue generated per day over time?
...

  2. When prompted, commit the generated semantic layer.

Install

Next Page

On this page

Run Databao CLISign inGenerate semantic layerReview and commit the semantic layer