Skip to content

Curriculum Curator CLI Reference

This document provides a reference for all available Curriculum Curator command-line interface (CLI) commands and options.

Command Overview

Command Description
interactive Launch interactive mode with a menu of common operations
run Run a specified workflow
resume Resume a previously interrupted workflow session
list-workflows List all available workflows
list-prompts List available prompts, optionally filtered by tag
list-validators List available content validators
list-remediators List available content remediators
build-workflow Interactive workflow builder for creating/editing workflow configurations
edit-prompt Interactive prompt editor for creating/editing prompt templates
init Initialize a new project with example configuration and prompts

General Options

These options are available in most commands:

Option Description
--config, -c Path to configuration file (default: config.yaml)
--help Show help message and exit

Command Details

Run a Workflow

Run a specified workflow with optional variables.

curator run [OPTIONS] WORKFLOW

Arguments: - WORKFLOW: Name of the workflow to run (required)

Options: - --var, -v: Variables in key=value format. Can be used multiple times - --session-id: Specify a session ID to use or resume - --config, -c: Path to configuration file (default: config.yaml) - --output-json, -j: Output result as JSON instead of rich text format

Examples: 

# Run a workflow with variables
curator run minimal_educational_module --var course_title="Python Programming" --var module_id="1"

# Run with custom configuration file
curator run standard_course --config custom_config.yaml

# Run with JSON output
curator run minimal_educational_module --output-json > output.json

Resume a Workflow

Resume a previously interrupted workflow session.

curator resume [OPTIONS] SESSION_ID

Arguments: - SESSION_ID: The Session ID of the workflow to resume (required)

Options: - --from-step: Specific step name to resume from - --config, -c: Path to configuration file (default: config.yaml) - --output-json, -j: Output result as JSON

Examples: 

# Resume a workflow session
curator resume 2023-04-05-12345-abc123

# Resume from a specific step
curator resume 2023-04-05-12345-abc123 --from-step validate_content

List Workflows

List all available workflows defined in the configuration file and predefined workflows.

curator list-workflows [OPTIONS]

Options: - --config, -c: Path to configuration file (default: config.yaml)

Example: 

curator list-workflows

List Prompts

List available prompts, optionally filtered by tag.

curator list-prompts [OPTIONS]

Options: - --tag, -t: Filter prompts by tag specified in YAML front matter - --config, -c: Path to configuration file (default: config.yaml)

Examples: 

# List all prompts
curator list-prompts

# List prompts with a specific tag
curator list-prompts --tag course

List Validators

List available content validators that can be used in workflows.

curator list-validators

Example: 

curator list-validators

List Remediators

List available content remediators that can be used in workflows.

curator list-remediators

Example: 

curator list-remediators

Build Workflow

Interactive workflow builder to create or edit workflow configurations.

curator build-workflow [OPTIONS] OUTPUT_FILE

Arguments: - OUTPUT_FILE: Path to save the workflow configuration (required)

Options: - --base, -b: Base workflow to start from - --config, -c: Path to configuration file (default: config.yaml)

Examples: 

# Create a new workflow from scratch
curator build-workflow my-workflow.yaml

# Create a workflow based on an existing one
curator build-workflow new-workflow.yaml --base existing-workflow.yaml

Edit Prompts

Interactive editor for creating and editing prompt templates.

curator edit-prompt [OPTIONS] [PROMPT_PATH]

Arguments: - PROMPT_PATH: Path to the prompt file to edit (optional)

Options: - --config, -c: Path to configuration file (default: config.yaml)

Examples: 

# Launch prompt editor menu
curator edit-prompt

# Edit a specific prompt
curator edit-prompt course/overview.txt

Interactive Mode

Launch interactive mode with a menu of common operations.

curator interactive [OPTIONS]

Options: - --config, -c: Path to configuration file (default: config.yaml)

Example: 

curator interactive

Initialize Project

Initialize a new project with example prompts and configuration.

curator init [OPTIONS] [OUTPUT_DIR]

Arguments: - OUTPUT_DIR: Directory to initialize with example prompts and configuration (default: current directory)

Example: 

# Initialize in current directory
curator init

# Initialize in a specific directory
curator init my-project-dir

Environment Variables

The Curriculum Curator CLI respects the following environment variables:

  • OPENAI_API_KEY: API key for OpenAI
  • ANTHROPIC_API_KEY: API key for Anthropic
  • GROQ_API_KEY: API key for Groq
  • GOOGLE_API_KEY: API key for Google AI (Gemini)

You can set these in your environment or in the configuration file.

Error Handling

When an error occurs, the CLI will: 1. Display an error message with details 2. Log additional information (if enabled) 3. Exit with a non-zero exit code

Example error handling: 

# If a workflow is not found
curator run non_existent_workflow
# [bold red]Error running workflow 'non_existent_workflow':[/bold red] Workflow not found

Logging

By default, logs are written to the console. You can configure logging behavior in the configuration file.

Configuration File

The configuration file (default: config.yaml) defines: - LLM providers and credentials - Base paths for prompts and output - Default settings for components - Custom workflow definitions

See the Configuration Guide for detailed information on the configuration file format.

Further Reading