Skip to content

Getting Started with the CLI

The bead CLI provides command-line tools for every stage of the experimental pipeline. This approach uses configuration files and shell commands to avoid Python programming.

Quick Start

# Verify bead CLI is available
uv run bead --help | head -5

When to Use the CLI

Use the CLI when: - You prefer configuration-driven workflows - You want to avoid Python programming - You're composing operations in shell scripts - You're working with single resources or templates - Your workflow is linear and straightforward

For batch operations, complex logic, or dynamic configuration, see the Python API.

Command Groups

The CLI organizes commands into groups by pipeline stage:

  • resources: Create lexicons and templates (Stage 1)
  • templates: Fill templates with lexical items (Stage 2)
  • items: Construct experimental items (Stage 3)
  • lists: Partition items into experiment lists (Stage 4)
  • deployment: Generate jsPsych/JATOS experiments (Stage 5)
  • training: Collect data and train models (Stage 6)
  • workflow: Run complete multi-stage pipelines
  • config: Manage configuration files

Getting Help

View available commands: 

uv run bead --help
uv run bead resources --help
uv run bead templates fill --help

Configuration Files

The CLI uses YAML configuration files to define pipeline parameters:

project:
  name: "my_experiment"
  language_code: "eng"

paths:
  lexicons_dir: "lexicons"
  templates_dir: "templates"
  items_dir: "items"

See Configuration Guide for complete reference.

Complete Workflow Example

For a complete working example using all 6 stages, see CLI Workflows.

Next Steps