Implementation:Guardrails ai Guardrails Configure CLI

Knowledge Sources	Guardrails Guardrails CLI Reference
Domains	Configuration, CLI
Last Updated	2026-02-14 00:00 GMT

Overview

Concrete CLI command for initializing Guardrails credentials and preferences provided by the guardrails package.

Description

The configure command is a Typer-based CLI command that prompts the user for or accepts Hub API tokens, metrics collection preferences, and remote inferencing settings. It writes these to a ~/.guardrailsrc file and validates the token against the Hub authentication service.

Usage

Run this command when setting up Guardrails for the first time, or when you need to update your Hub API key or change metrics/inference preferences.

Code Reference

Source Location

• Repository: guardrails
• File: guardrails/cli/configure.py
• Lines: L57-140

Signature

@guardrails.command()
def configure(
    enable_metrics: Optional[bool] = typer.Option(
        True,
        "--enable-metrics/--disable-metrics",
        help="Opt out of anonymous metrics collection.",
        prompt="Enable anonymous metrics reporting?",
    ),
    token: Optional[str] = typer.Option(
        None,
        "--token",
        help="API Key for Guardrails.",
    ),
    remote_inferencing: Optional[bool] = typer.Option(
        True,
        "--enable-remote-inferencing/--disable-remote-inferencing",
        help="Opt in to remote inferencing.",
        prompt="Do you wish to use remote inferencing?",
    ),
    clear_token: Optional[bool] = typer.Option(
        False,
        "--clear-token",
        help="Clear the existing token from the configuration file.",
    ),
):

Import

# CLI command - invoked via shell:
# guardrails configure

I/O Contract

Inputs

Name	Type	Required	Description
enable_metrics	Optional[bool]	No	Opt in/out of anonymous metrics collection (default True)
token	Optional[str]	No	API Key for Guardrails Hub
remote_inferencing	Optional[bool]	No	Enable remote ML inference for validators (default True)
clear_token	Optional[bool]	No	Clear existing token from configuration file

Outputs

Name	Type	Description
~/.guardrailsrc	File	Configuration file with id, token, enable_metrics, use_remote_inferencing
settings singleton	Settings	Updated global settings singleton with new configuration

Usage Examples

Interactive Configuration

# Run with interactive prompts
guardrails configure

# Non-interactive with flags
guardrails configure --token "my-api-key" --enable-metrics --enable-remote-inferencing

Clear Token

# Clear stored token
guardrails configure --clear-token

Related Pages

Implements Principle

• Principle:Guardrails_ai_Guardrails_CLI_Configuration

Requires Environment

• Environment:Guardrails_ai_Guardrails_LLM_Provider_API_Keys