Product safety teams at companies deploying LLMs in regulated or consumer-facing contexts need to translate high-level content policies into precise, testable guardrail specifications that engineering teams can implement and audit. This agent converts policy intent into structured, implementable guardrail rules with test cases.