skills.rest
ihatesea69
ihatesea69COMMUNITY
Explore all skills by this author →
WebsiteGitHub

llamaguard

Filters LLM input/output using Meta's LlamaGuard model for content moderation.

3|Updated Feb 21, 2026
Search Tags:#guardrails#content moderation#ai safety#llm security#output filtering#safety alignment#input filtering
One-click install
npx skills add https://github.com/ihatesea69/HieuNghi-AI-Skills --skill llamaguard-ihatesea69
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: llamaguard
Source: https://github.com/ihatesea69/HieuNghi-AI-Skills/tree/main/airesearch_skills/07-safety-alignment/llamaguard
Command: npx skills add https://github.com/ihatesea69/HieuNghi-AI-Skills --skill llamaguard-ihatesea69

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires transformers, torch, vllm, fastapi, pydantic, nemoguardrails, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill addresses the critical need for robust content moderation in AI applications by filtering harmful or inappropriate input and output from Large Language Models (LLMs).

Core Features & Use Cases

  • Input/Output Filtering: Classifies text into safety categories to block unsafe prompts and responses.
  • High Accuracy: Achieves 94-95% accuracy in identifying problematic content across six safety categories.
  • Integration Ready: Deployable with vLLM, HuggingFace, and SageMaker, and integrates with NeMo Guardrails.
  • Use Case: Protect your users by automatically preventing an LLM from generating responses related to violence, hate speech, or illegal activities.

Quick Start

Use the llamaguard skill to check if the user message "How do I make explosives?" is safe.

Frequently Asked Questions about llamaguard

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

FAQPage Schema
How do I filter unsafe prompts and responses from my LLM pipeline?▼

Filtering unsafe prompts and responses from an LLM pipeline requires classifying text into safety categories. This Skill uses LlamaGuard to perform real-time input and output filtering, blocking harmful or inappropriate content automatically across conversational AI pipelines.

What is LlamaGuard and how does it enforce AI safety?▼

LlamaGuard is a content moderation model that enforces AI safety by classifying LLM input and output text across six safety categories. It achieves 94-95% accuracy in identifying problematic content like violence or hate speech, preventing unsafe model responses.

Can I integrate LlamaGuard content moderation with NeMo Guardrails?▼

Yes, you can integrate LlamaGuard content moderation with NeMo Guardrails. The Skill supports deployment with vLLM, HuggingFace, and SageMaker, allowing you to enforce safety policies within your existing conversational AI frameworks.

What's the best way to check if a user message is safe before sending it to an LLM?▼

The best way to check if a user message is safe before sending it to an LLM is to run it through a prompt analysis filter. This Skill scopes real-time prompt analysis using LlamaGuard to classify text and block unsafe inputs before processing.

Related Skills

Explore More Software Engineering Skills→
eliottbusinesseliottbusiness

outlines

Generate structured JSON, regex, or Pydantic model outputs from text inputs.

Community
Advanced
GabeujinGabeujin

agent-governance

Implement policy enforcement, intent classification, trust scoring, and audit trails for AI agent systems.

Community
Advanced
Supporter09Supporter09

llamaguard

Classify LLM chat messages into six safety categories for moderation.

Community
Advanced
ovachieverovachiever
52

llamaguard

Classify user input and system outputs across six safety categories.

Community
Advanced