Agentic AI Reference Architecture

Safety & Security · Software component

LLM Self-Check Output Rail

Software componentSafety & SecuritySafety, Security & Governancearc:LLMSelfCheckOutputRail

An output guardrail that prompts the LLM itself, as a judge, to assess whether its own proposed output violates content policies before delivery.

Responsibility. Screens outputs for policy violations using the LLM as judge.

Also known as: self check output flow, LLM-as-judge output check

guards; invokesis orchestrated byLLM Inference Service: guards; invokesLLM Inference ServiceGuardrail Orchestrator: is orchestrated byGuardrail Orchestrator
Direct neighbourhood (hover for relationship types)

Relationships

invokes dependency

guards control

is orchestrated by control

Classification

Patterns
LLM-as-judge
Technologies
NeMo Guardrails
Risks mitigated
Common toxic or harmful content

Sources

  1. Ch9.1: T. Nguyen, "Output Filtering and Content Moderation," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 9.1. ISBN: 9798244538229.