skills.rest
Clay-HHK
Clay-HHKCOMMUNITY
Explore all skills by this author →
GitHub

nemo-guardrails

Add programmable runtime safety rails to LLM interactions with Colang-defined flows.

2|Updated Apr 12, 2026
Search Tags:#safety#guardrails#colang#hallucination-detection#llm-safety#jailbreak-detection#pii-filtering
One-click install
npx skills add https://github.com/Clay-HHK/claude-config --skill nemo-guardrails-clay-hhk
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: nemo-guardrails
Source: https://github.com/Clay-HHK/claude-config/tree/main/skills/AI-research-SKILLs/07-safety-alignment/nemo-guardrails
Command: npx skills add https://github.com/Clay-HHK/claude-config --skill nemo-guardrails-clay-hhk

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

NeMo Guardrails provides runtime safety controls for large language model applications to detect and block jailbreaks, filter sensitive data, reduce hallucinations, and enforce input/output validation so applications behave reliably and comply with safety policies.

Core Features & Use Cases

  • Jailbreak & prompt-injection detection: Pattern matching and LLM-assisted checks to block malicious or adversarial user inputs before they reach the model.
  • Input/output self-checks and fact-checking: Custom actions and retrieval integrations validate user inputs and verify model outputs to reduce hallucinations.
  • PII and toxicity filtering: Integrations with Presidio, ActiveFence, and moderation models like LlamaGuard for masking or blocking sensitive content.
  • Production readiness: Colang 2.0 DSL for programmable flows, low-latency checks, and GPU-accelerated deployment options for enterprise LLM services.

Quick Start

Install nemoguardrails and initialize LLMRails with a RailsConfig that defines flows for jailbreak prevention, PII masking, and fact-checking to protect your application at runtime.

Frequently Asked Questions about nemo-guardrails

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I add runtime guardrails to LLM applications to detect jailbreaks?▼

To add runtime guardrails to LLM applications for jailbreak detection, you initialize a rails configuration with Colang flows that apply pattern matching and LLM-assisted checks to block malicious inputs before they reach the model.

What is the best way to filter PII in LLM interactions at runtime?▼

The best way to filter PII in LLM interactions is by integrating custom actions like Presidio into your runtime guardrails, which masks or blocks sensitive content during input and output validation.

Does NeMo Guardrails work with retrieval-augmented generation systems for fact-checking?▼

Yes, NeMo Guardrails works with retrieval-augmented generation systems by using custom actions and retrieval integrations to validate user inputs and verify model outputs, which reduces hallucinations.

Can I use LlamaGuard for toxicity filtering in production LLM services?▼

You can use LlamaGuard for toxicity filtering in production LLM services by defining it as a custom action integration within your programmable guardrails configuration to block sensitive content.

Do I need a GPU for low-latency safety checks in LLM moderation pipelines?▼

You do not strictly need a GPU for low-latency safety checks in LLM moderation pipelines, but GPU-accelerated deployment options are supported for enterprise services requiring high throughput.

How do I define custom flows for input and output validation in Colang?▼

You define custom flows for input and output validation in Colang by using the 2.0 DSL to programmatically specify guardrail behaviors, integrating external tools to enforce application safety policies.

Related Skills

Explore More Software Engineering Skills→
pluginagentmarketplacepluginagentmarketplace
3

input-output-guardrails

Enforce input and output guardrails on AI interactions.

Community
Basic
ovachieverovachiever
52

tensorrt-llm

Optimize LLM inference on NVIDIA GPUs with TensorRT-LLM.

Community
Advanced
ovachieverovachiever
52

nemo-guardrails

Enforce runtime safety rails for LLM applications using Colang 2.0 DSL.

Community
Advanced
tadod12tadod12

nemo-guardrails

Configure programmable runtime safety rails for LLM applications using Colang 2.0 DSL.

Community
Advanced