nemo-guardrails

Implement runtime safety for LLM applications with NVIDIA NeMo Guardrails.

3|Updated Feb 21, 2026

Search Tags:#safety #hallucination detection #prompt injection #guardrails #colang #pii filtering #llm security

Ready to run

npx skills add https://github.com/ihatesea69/HieuNghi-AI-Skills --skill nemo-guardrails-ihatesea69

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: nemo-guardrails
Source: https://github.com/ihatesea69/HieuNghi-AI-Skills/tree/main/airesearch_skills/07-safety-alignment/nemo-guardrails
Command: npx skills add https://github.com/ihatesea69/HieuNghi-AI-Skills --skill nemo-guardrails-ihatesea69

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill provides a robust framework for adding runtime safety and security to Large Language Model (LLM) applications, preventing misuse and ensuring reliable outputs.

Core Features & Use Cases

Jailbreak Detection: Prevents prompt injection attacks and unauthorized behavior modifications.
Input/Output Validation: Ensures user inputs and LLM outputs adhere to defined safety policies.
Fact-Checking & Hallucination Detection: Verifies factual accuracy and reduces the generation of false information.
PII Filtering: Protects sensitive user data by detecting and masking Personally Identifiable Information.
Toxicity Detection: Filters out harmful or inappropriate content.
Use Case: Integrate NeMo Guardrails into a customer service chatbot to automatically detect and block requests for illegal activities, filter out toxic language, and ensure the bot doesn't hallucinate product information.

Quick Start

Install the library and use the LLMRails class to wrap your LLM with basic input validation rules.

Frequently Asked Questions about nemo-guardrails

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

FAQPage Schema

How do I prevent prompt injection attacks in my LLM application?▼

Prevent prompt injection attacks by implementing runtime guardrails that detect jailbreaks and validate inputs against defined safety policies using a Colang 2.0 DSL. This blocks unauthorized behavior modifications before they reach the model.

How do I filter PII from LLM outputs?▼

Filter PII from LLM outputs by applying runtime safety rules that detect and mask Personally Identifiable Information. This protects sensitive user data by ensuring it never appears in the final generated response.

What is the best way to detect hallucinations in LLM responses?▼

Detect hallucinations in LLM responses by applying runtime fact-checking guardrails. This verifies factual accuracy during generation and reduces the output of false information without modifying the core model.

Can I use Colang to define input and output validation for chatbots?▼

Use Colang 2.0 to programmatically define input and output validation rules for chatbots. This ensures both user inputs and LLM outputs strictly adhere to your defined safety policies.

Does runtime LLM security support production deployments with GPU acceleration?▼

Runtime LLM security supports production deployments with GPU acceleration. This allows you to scale safety mechanisms like toxicity detection and jailbreak prevention in high-traffic environments.

Why does my customer service chatbot generate inappropriate content?▼

Customer service chatbots generate inappropriate content when lacking runtime toxicity detection. Implementing guardrails filters harmful language and blocks unsafe requests, ensuring outputs remain safe and reliable.