Creates a guardrail to detect and filter harmful content in your generative AI application.