Jump to content

Connect SuperML | Leeroopedia MCP: Equip your AI agents with best practices, code verification, and debugging knowledge. Powered by Leeroo — building Organizational Superintelligence. Contact us at founders@leeroo.com.

Implementation:Guardrails ai Guardrails Create Command Hub

From Leeroopedia
Knowledge Sources
Domains Package_Distribution, CLI
Last Updated 2026-02-14 00:00 GMT

Overview

Concrete CLI command for scaffolding Guard configurations from Hub validators provided by the guardrails package.

Description

The guardrails create command (create_command function) installs specified validators from the Hub and generates a config.py file with proper imports, Guard instantiation, and .use() calls. It supports templates for pre-built configurations and a dry-run mode for previewing output. This is an External Tool Doc as it documents a CLI command rather than a programmatic API.

Usage

Run guardrails create --validators "hub://guardrails/regex_match,hub://guardrails/toxic_language" --guard-name MyGuard to generate a config file.

Code Reference

Source Location

  • Repository: guardrails
  • File: guardrails/cli/create.py
  • Lines: L18-104

Signature

@gr_cli.command(name="create")
@trace(name="guardrails-cli/create")
def create_command(
    validators: Optional[str] = typer.Option(
        default="", help="A comma-separated list of validator hub URIs.",
    ),
    guard_name: Optional[str] = typer.Option(
        default=None, help="The name of the guard to define in the file.",
    ),
    local_models: Optional[bool] = typer.Option(
        None, "--install-local-models/--no-install-local-models",
    ),
    filepath: str = typer.Option(
        default="config.py", help="Output configuration file path.",
    ),
    template: Optional[str] = typer.Option(
        default=None, help="Hub template URI for pre-built configurations.",
    ),
    dry_run: bool = typer.Option(
        default=False, is_flag=True, help="Preview without writing.",
    ),
):

Import

# CLI command - invoked via shell:
# guardrails create --validators "hub://guardrails/regex_match"

I/O Contract

Inputs

Name Type Required Description
validators Optional[str] No Comma-separated Hub URIs
guard_name Optional[str] No Name for the generated Guard
filepath str No Output path (default "config.py")
template Optional[str] No Hub template URI for pre-built configs
dry_run bool No Preview without writing (default False)

Outputs

Name Type Description
config.py File Python file with validator imports, Guard instantiation, and .use() calls

Usage Examples

Basic Scaffolding

# Generate a config with two validators
guardrails create \
    --validators "hub://guardrails/regex_match,hub://guardrails/toxic_language" \
    --guard-name ContentGuard

# Preview without writing
guardrails create \
    --validators "hub://guardrails/regex_match" \
    --dry-run

Using Templates

# Use a pre-built chatbot template
guardrails create --template hub:template://guardrails/chatbot

Related Pages

Implements Principle

Page Connections

Double-click a node to navigate. Hold to expand connections.
Principle
Implementation
Heuristic
Environment