Usage

Usage Guide for rai-checklist-cli

1. Installation

Install the Responsible AI Checklist CLI using pip:

pip install rai-checklist-cli

Note: PyYAML is a dependency. You may need to install it separately if you encounter issues.

2. Basic Usage

The basic syntax for using the CLI is:

rai-checklist [OPTIONS]

To generate a default checklist:

rai-checklist generate

3. Command-Line Options

• -h, --help: Show help message and exit
• -w, --overwrite: Overwrite existing output file
• -o, --output PATH: Specify output file path
• -f, --format TEXT: Specify output format (md, yaml, json)
• -l, --checklist PATH: Path to custom checklist file
• --project-type TEXT: Specify project type for validation (default, machine_learning, web_application, etc.)
• --config PATH: Path to the configuration file for validation

4. Output Formats

The tool supports three output formats:

• Markdown (md)
• YAML
• JSON

Specify the format using the -f or --format option:

rai-checklist generate -f yaml

5. Custom Templates

To create a custom template:

rai-checklist create-template

Follow the prompts to name your template and add sections and items.

To use a custom template:

rai-checklist generate -t your_custom_template

6. Validation

To validate a checklist against project requirements:

rai-checklist validate path/to/your/checklist.yaml --project-type machine_learning

7. CI/CD Integration

The tool can be integrated into CI/CD pipelines. An example GitHub Actions workflow is provided in the repository.

8. Jupyter Notebook Usage

The tool can be used within Jupyter notebooks. Import the necessary modules:

from rai_checklist_cli.checklist_generator import generate_checklist
import yaml

Then use the generate_checklist function to create checklists programmatically.

9. Configuration

The tool uses a configuration file located at ~/.rai_checklist_config.yaml. You can specify custom paths or settings in this file.

10. Troubleshooting

If you encounter issues:

• Ensure you have the latest version installed
• Check that all dependencies are correctly installed
• Verify that your custom templates or configuration files are correctly formatted

For more help, please open an issue on the GitHub repository.