Commands Reference

Complete reference for all dataspoc-lens CLI commands and their options.

Global options

dataspoc-lens --version    # Show version
dataspoc-lens --help       # Show help

init

Initialize DataSpoc Lens configuration.

dataspoc-lens init

Creates ~/.dataspoc-lens/ directory with config.yaml and transforms/ subdirectory.

add-bucket

Register a bucket and discover tables.

dataspoc-lens add-bucket <uri>

Argument	Description
uri	Bucket URI: s3://, gs://, az://, or file://

Reads the Pipe manifest or scans for .parquet files and prints discovered tables.

catalog

List all tables from registered buckets.

dataspoc-lens catalog [OPTIONS]

Option	Default	Description
--detail <table>	—	Show detailed schema for a specific table
--output <format>	table	Output format: table or json

Examples:

dataspoc-lens catalog
dataspoc-lens catalog --detail orders
dataspoc-lens catalog --output json

query

Execute a SQL query and print results.

dataspoc-lens query <sql> [OPTIONS]

Argument / Option	Default	Description
sql	—	SQL query to execute
--export, -e	—	Export results to file (format from extension: .csv, .json, .parquet)
--output	table	Output format: table or json

Examples:

dataspoc-lens query "SELECT * FROM orders LIMIT 10"
dataspoc-lens query "SELECT * FROM orders" --export results.csv
dataspoc-lens query "SELECT COUNT(*) FROM orders" --output json

shell

Launch the interactive SQL shell.

dataspoc-lens shell

No options. See Interactive Shell for details on dot commands and features.

ask

Ask a question in natural language and get SQL results.

dataspoc-lens ask <question> [OPTIONS]

Argument / Option	Default	Description
question	—	Natural language question
--debug	false	Show the prompt sent to the LLM
--export, -e	—	Export results to file (.csv, .json, .parquet)
--output	table	Output format: table or json

Examples:

dataspoc-lens ask "What are the top 10 customers by revenue?"
dataspoc-lens ask "Monthly order trends" --debug
dataspoc-lens ask "Average order value by status" --export report.csv
dataspoc-lens ask "Total revenue" --output json

setup-ai

Install and configure Ollama for local AI (free, no API key needed).

dataspoc-lens setup-ai

notebook

Open an interactive notebook with tables pre-mounted.

dataspoc-lens notebook [OPTIONS]

Option	Default	Description
--marimo	false	Use Marimo instead of JupyterLab

Examples:

dataspoc-lens notebook
dataspoc-lens notebook --marimo

transform run

Run SQL transforms in order.

dataspoc-lens transform run

Executes all numbered .sql files in ~/.dataspoc-lens/transforms/ in order. See SQL Transforms for details.

transform list

List available transform files.

dataspoc-lens transform list

cache

Manage local cache of remote Parquet data.

dataspoc-lens cache [TABLE] [OPTIONS]

Argument / Option	Default	Description
table	—	Table name to cache
--list	false	List cached tables
--refresh	false	Force re-download of cached data
--clear	false	Clear cached data (all or specific table)
--output	table	Output format: table or json

Examples:

dataspoc-lens cache orders                # Cache the orders table
dataspoc-lens cache --list                # List all cached tables
dataspoc-lens cache --list --output json  # List cached tables as JSON
dataspoc-lens cache orders --refresh      # Force re-download
dataspoc-lens cache --clear               # Clear all cached data
dataspoc-lens cache orders --clear        # Clear specific table cache

mcp

Start the MCP server for AI agent integration.

dataspoc-lens mcp

See MCP Server for configuration details.

ml activate

Activate DataSpoc ML with a license key.

dataspoc-lens ml activate [KEY]

Without a key, displays information about DataSpoc ML and how to purchase a license.

ml status

Show DataSpoc ML license status.

dataspoc-lens ml status

ml train

Train a model via DataSpoc ML.

dataspoc-lens ml train --target <column> --from <table>

Option	Description
--target	Target column name for prediction
--from	Table name or path to training data

ml predict

Generate predictions via DataSpoc ML.

dataspoc-lens ml predict --model <name> --from <table>

Option	Description
--model	Model name to use for predictions
--from	Table name or path to input data

ml models

List trained ML models.

dataspoc-lens ml models

ml explain

Explain a trained model via DataSpoc ML.

dataspoc-lens ml explain --model <name>

Option	Description
--model	Model name to explain