gws_core

GWS CLI Documentation

Overview

GWS CLI is a command-line interface for managing the Genestack Workspace (GWS) environment, including server operations, brick management, task generation, and application development.

Installation

gws [COMMAND] [OPTIONS]

Global Options

• --log-level [INFO|DEBUG|ERROR] - Set global logging level for all commands (default: INFO)
• -h, --help - Show help message

Commands

server - Server Operations

Manage server operations including running the server, testing, and executing scenarios/processes.

server run

Start the development server.

Options:

• --port TEXT - Server port (default: "3000")
• --settings-path TEXT - Path to main settings file
• --show-sql - Log SQL queries in console (flag)
• --allow-dev-app-connections - Allow connections from apps running in dev mode (flag, dev mode only)

Example:

gws server run --port 3000 --log-level DEBUG

server test

Run tests for a specific brick or all bricks.

Arguments:

• TEST_NAME - Test file name(s) to launch (regex) or 'all' for all tests

Options:

• --brick-name TEXT - Brick name to test (uses current folder if omitted)
• --show-sql - Log SQL queries (flag)

Example:

gws server test test_table_copilot
gws server test all --brick-name gws_ai_toolkit

server run-scenario

Execute a specific scenario by ID.

Options:

• --scenario-id TEXT - Scenario ID (required)
• --user-id TEXT - User ID that runs the scenario (required)
• --settings-path TEXT - Path to main settings file
• --show-sql - Log SQL queries (flag)
• --test - Run in test mode (flag)

server run-process

Execute a specific process within a scenario.

Options:

• --scenario-id TEXT - Scenario ID (required)
• --protocol-model-id TEXT - Protocol model ID (required)
• --process-instance-name TEXT - Process instance name (required)
• --user-id TEXT - User ID (required)
• --settings-path TEXT - Path to main settings file
• --show-sql - Log SQL queries (flag)
• --test - Run in test mode (flag)

brick - Brick Management

Generate and manage bricks (reusable data processing components).

brick generate

Generate a new brick with boilerplate code and structure.

Arguments:

• NAME - Brick name (snake_case recommended)

Example:

gws brick generate my_custom_brick

brick install-deps

Install pip dependencies from a brick's settings.json file.

Arguments:

• SETTINGS_PATH - Path to settings.json file

Example:

gws brick install-deps ./bricks/my_brick/settings.json

task - Task Generation

Generate task classes for data processing workflows.

task generate

Generate a new task class with boilerplate code.

Arguments:

• NAME - Task class name (PascalCase)

Options:

• --human-name TEXT - Human-readable task name
• --short-description TEXT - Short description of the task

Example:

gws task generate MyDataProcessor --human-name "Data Processor" --short-description "Processes data files"

streamlit - Streamlit Applications

Generate and run Streamlit applications.

streamlit run

Run a Streamlit app in development mode.

Arguments:

• CONFIG_FILE_PATH - Path to JSON config file

Options:

• --enable-debugger - Enable debugger in Streamlit app (flag)

Example:

gws streamlit run ./config.json --enable-debugger

streamlit generate

Generate a new Streamlit app with boilerplate code.

Arguments:

• NAME - App name (snake_case)

Example:

gws streamlit generate my_dashboard

reflex - Reflex Applications

Generate and run Reflex applications.

reflex run

Run a Reflex app in development mode.

Arguments:

• CONFIG_FILE_PATH - Path to JSON config file

Example:

gws reflex run ./dev-config.json

reflex generate

Generate a new Reflex app.

Arguments:

• NAME - App name (snake_case)

Options:

• --enterprise - Generate enterprise Reflex app (flag)

Example:

gws reflex generate my_app --enterprise

reflex init

Alias for reflex generate.

dev-env - Development Environment

Manage development environment data and settings.

dev-env reset

Reset development environment data (WARNING: irreversible).

Interactive: Prompts for confirmation before deleting data.

Example:

gws dev-env reset

claude - Claude Code Integration

Manage Claude Code integration and configuration.

claude install

Install Claude Code CLI tool (automatically installs Node.js if needed).

Example:

gws claude install

claude update

Update Claude Code configuration for GWS (commands and settings). Only runs if Claude Code is already installed.

Example:

gws claude update

claude commands

Manage Claude Code commands.

Options:

• --pull - Copy GWS commands to ~/.claude/commands/gws-commands (flag)
• --list - List all available GWS commands (flag)

Example:

gws claude commands --pull
gws claude commands --list

copilot - GitHub Copilot Integration

Manage GitHub Copilot integration and configuration.

copilot update

Update GitHub Copilot configuration for GWS (instructions and settings).

Example:

gws copilot update

copilot instructions

Manage GitHub Copilot instructions.

Options:

• --pull - Copy GWS instructions to ~/.github/copilot/instructions/gws-instructions (flag)
• --list - List all available GWS commands (flag)

Example:

gws copilot instructions --pull
gws copilot instructions --list

copilot commands

Alias for copilot instructions.

utils - Utility Commands

Utility commands for development environment setup.

utils install node

Install Node.js using NVM (Node Version Manager).

Example:

gws utils install node

utils screenshot

Take a screenshot of a web application using Playwright.

Options:

• --url, -u TEXT - Base URL of the application (default: "http://localhost:8511")
• --route, -r TEXT - Route to navigate to (default: "/")
• --output, -o TEXT - Output path for screenshot (default: /lab/user/app_screenshot.png)
• --no-logs - Don't save console logs (flag)
• --headless/--no-headless - Run browser in headless mode (default: headless)

Example:

gws utils screenshot --route /dashboard --output ./screenshots/dashboard.png
gws utils screenshot --url http://localhost:3000 --no-headless