databao 0.6.1.dev1

QA agent for the Databao platform

Databao CLI

Databao is a data assistant in Slack for all your teams, with answers you can trust.

Databao CLI is the local toolkit for building and maintaining the semantic layer that powers the assistant for dbt projects.

Databao • Quickstart • Docs

---

Prerequisites

1. Python 3.11+
2. uv
3. Claude Code
4. Google account
5. Git repository with your dbt project

---

Installation

Databao CLI is available on PyPI.

Install globally

uv tool install databao

Run without installing

uvx databao

---

Quickstart

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

2. Launch Databao:

   databao
   

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

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

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

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

6. When the semantic layer is generated, Claude will offer you to commit and push it to your GitHub repository.

For more details, check out the quickstart guide in the Databao docs.

---

Troubleshooting

Running Docker on ARM Macs (Apple Silicon)

The Kaleido container used for plot rendering relies on Chromium, which does not run natively on ARM. To make it work, enable Rosetta emulation in Docker Desktop:

1. Open Docker Desktop -> Settings -> General.
2. Enable Use Rosetta for x86_64/amd64 emulation on Apple Silicon.
3. Click Apply & Restart.

Without this, the Kaleido/Chromium container will fail to start on ARM Macs. Docker's default QEMU emulation doesn't cover all the x86 instructions Chromium uses and crashes on them (illegal-instruction / SIGILL), so Rosetta is required here.

---

Databao • Docs