CLI commands
Get familiar with CLI commands and options for Databao
Basic usage:
databao <command> [OPTIONS]Global options
| Flag | Description |
|---|---|
-p, --project-dir PATH | Location of your Databao project. Default is the current working directory. |
CLI commands
| Command | Description | Example |
|---|---|---|
status | Display project status and system-wide information. | databao status |
index | Re-index contexts without rebuilding them. | databao index |
init | Create a new Databao project. | databao init |
build | Build context for all domain's data sources. | databao build |
ask | Chat with the Databao agent. | databao ask |
app | Open the Databao app in the browser. | databao app |
status
Displays project status and system-wide information.
databao status [OPTIONS]index
Update the semantic index for already-built contexts (derived from your data sources) without triggering a full rebuild.
You can use this command to update the semantic index of your data sources without going through the time- and resource-intensive rebuild process.
databao index [OPTIONS] [DATASOURCES_CONFIG_FILES]...| Option | Description |
|---|---|
-d, --domain TEXT | Databao domain name. Default is root. |
By default, all contexts are re-indexed. To target specific data sources, provide one or more config file names.
Examples:
# Re-index all contexts in the default domain (root)
databao index
# Re-index all contexts in a specific domain
databao index -d my_domain
# Re-index the specified data sources
databao index snowflake.yamlConfig file names must include the file extension, such as snowflake.yaml.
You can find data source config files at /domains/<domain_name>/src in your Databao project directory.
init
Create a new Databao project in the current directory or in the specified directory.
It creates the project structure (.databao/domains/root, .databao/agents)
and allows you to configure data sources interactively.
databao init [OPTIONS]build
Build context for all domain's data sources. The output is saved in the domain's output directory.
databao build [OPTIONS]| Option | Description |
|---|---|
-d, --domain TEXT | Databao domain name. Default is "root". |
ask
Chat with the Databao agent.
The command starts an interactive chat session with the Databao agent or executes a single query if --one-shot is provided.
databao ask [OPTIONS] [QUESTION]| Option | Description |
|---|---|
--one-shot | Execute a single question and exit. Default is false. If not provided, inteactive mode is started. |
-m, --model TEXT | LLM model in the provider:name format. Examples: openai:gpt-4o, anthropic:claude-sonnet-4-5. |
-t, --temperature FLOAT | Temperature in the 0.0-1.0 range. Default is 0.0. |
--show-thinking / --no-show-thinking | Display reasoning/thinking output. Default is true. |
Examples:
databao ask # Interactive mode
databao ask --one-shot "What tables exist?" # One-shot mode
databao ask --model anthropic:claude-sonnet-4-5 # Use custom model
databao ask --no-show-thinking # Hide reasoningapp
Launch Databao in the browser. You can initialize a new project, add data sources, build context, and create chats in the UI.
databao app [OPTIONS] [STREAMLIT_ARGS]...Examples:
databao app
databao app --server.port 8502
databao app --server.headless true