skills.rest
chicagopeabodydev-sudo
chicagopeabodydev-sudoCOMMUNITY
Explore all skills by this author →
GitHub

nemo-guardrails-skill

Apply programmable input, retrieval, and output rails to LLM calls.

Updated Feb 11, 2026
Search Tags:#guardrails#open-source#llm-safety#custom-actions#retrieval-augmentation#content-filtering
One-click install
npx skills add https://github.com/chicagopeabodydev-sudo/library_bot_poc --skill nemo-guardrails-skill
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: nemo-guardrails-skill
Source: https://github.com/chicagopeabodydev-sudo/library_bot_poc/tree/main/.cursor/skills/nemo-guardrails-skill
Command: npx skills add https://github.com/chicagopeabodydev-sudo/library_bot_poc --skill nemo-guardrails-skill

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Guardrails provide programmable safety nets for LLM-driven chat, ensuring user inputs, retrieved content, and tool usage comply with policy and safety constraints.

Core Features & Use Cases

  • Input rails to block unsafe or off-topic questions before processing.
  • Retrieval rails to validate or filter context retrieved by RAG before it's fed to the model.
  • Output rails to sanitize or constrain the final answer before delivery.
  • Support for custom actions (e.g., in actions.py) to implement project-specific checks and flows.
  • Works across UI layers and back-end pipelines, enabling safe, reliable assistant experiences.

Quick Start

Install the nemoguardrails package, load a RailsConfig from a guardrails/ directory, and integrate LLMRails into your application to start applying input, retrieval, and output rails.

Frequently Asked Questions about nemo-guardrails-skill

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

FAQPage Schema
How do I add guardrails to LLM calls for content filtering and input validation?▼

To add guardrails to LLM calls, load a RailsConfig from a guardrails directory and integrate LLMRails into your application to apply programmable input, retrieval, and output rails for content filtering and input validation.

Can I filter retrieved RAG context before feeding it to the model?▼

Yes, you can filter retrieved RAG context by applying retrieval rails, which validate and filter the context retrieved by RAG pipelines before it is fed to the model for processing.

How do I sanitize or constrain LLM outputs before delivering them to users?▼

You can sanitize or constrain LLM outputs by applying output rails, which intercept and filter the final answer generated by the model before it is delivered to the user in chat UIs or pipelines.

How do I implement custom validation checks for LLM safety guardrails?▼

You can implement custom validation checks by defining project-specific logic in an optional actions.py file, extending the control flows and validation rules of your guardrails configuration.

What do I need to set up programmable guardrails for safe LLM conversations?▼

You need a Python environment with the nemoguardrails package installed, a guardrails configuration directory to load your RailsConfig, and optional custom actions to start applying programmable guardrails for safe LLM conversations.

Related Skills

Explore More Software Engineering Skills→
pluginagentmarketplacepluginagentmarketplace
3

input-output-guardrails

Enforce input and output guardrails on AI interactions.

Community
Basic
ConnectiveTCSConnectiveTCS

text-to-speech

Convert input text into Kokoro TTS audio with per-voice configuration.

Community
Advanced
ToanPV90ToanPV90

prompt-injection-defense

Sanitize inputs, validate tool calls, and redact leaked secrets in LLM pipelines.

Community
Advanced
ovachieverovachiever
52

nemo-guardrails

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

Community
Advanced