nemo-guardrails

Apply programmable runtime safety rails to LLM applications with Colang 2.0.

1.0k|117|Updated Feb 26, 2026

Search Tags:#llm #nemoguardrails #colang #hallucination-detection #jailbreak-detection #runtime-safety #pii-filtering

Ready to run

npx skills add https://github.com/OpenLAIR/dr-claw --skill nemo-guardrails-openlair

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: nemo-guardrails
Source: https://github.com/OpenLAIR/dr-claw/tree/main/skills/safety-alignment/nemo-guardrails
Command: npx skills add https://github.com/OpenLAIR/dr-claw --skill nemo-guardrails-openlair

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

NeMo Guardrails provides programmable runtime safety rails for LLM applications, reducing the risk of unsafe prompts, data leakage, and harmful outputs by enforcing policy-based checks at runtime.

Core Features & Use Cases

Jailbreak detection: identify and block prompt injection attempts that try to bypass safeguards.
Input/Output validation: enforce policy-compliant conversations by validating user inputs and model outputs.
Fact-checking and retrieval: verify factual claims and optionally fetch authoritative information.
PII filtering: redact sensitive information from prompts and responses before processing or logging.
Production-ready with Colang 2.0: programmable rails that run on GPUs and integrate with existing LLM pipelines.

Quick Start

Install nemoguardrails and configure a basic guardrails flow to block illegal prompts.

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 safety rails to an LLM application?▼

Runtime safety rails are added to LLM applications by applying programmable input/output validation, jailbreak detection, and PII filtering at runtime. This enforces policy-based checks to reduce the risk of unsafe prompts, data leakage, and harmful outputs.

What is the best way to detect jailbreak and prompt injection attempts in production LLMs?▼

Detecting jailbreak attempts in production LLMs is best handled by programmable runtime safety rails that identify and block prompt injection efforts. This approach prevents malicious actors from bypassing established model safeguards.

How does PII filtering work for large language model prompts and responses?▼

PII filtering for large language model prompts and responses works by redacting sensitive information before processing or logging occurs. Runtime safety rails enforce this programmatically, ensuring personally identifiable information is stripped from production conversations.

Do I need a GPU to run Colang 2.0 guardrails for hallucination detection?▼

Yes, you need a GPU to run Colang 2.0 guardrails for hallucination detection reliably. The runtime safety rails require GPU-accelerated environments, such as NVIDIA T4 setups, to process fact-checking and validation checks.

Can I perform fact-checking and retrieval validation on LLM outputs at runtime?▼

Yes, you can perform fact-checking and retrieval validation on LLM outputs at runtime. Programmable safety rails verify factual claims and optionally fetch authoritative information to ensure production responses are accurate.

What are the limitations of using programmable safety rails for LLM input validation?▼

Limitations of programmable safety rails for LLM input validation include the strict requirement for GPU-accelerated environments. Without adequate hardware like NVIDIA T4 GPUs, the runtime fact-checking and jailbreak detection processes cannot operate reliably.