nemo-guardrails

What problem does it solve?

This Skill provides a robust framework to ensure the safety, reliability, and ethical behavior of Large Language Model (LLM) applications at runtime.

Core Features & Use Cases

• Runtime Safety: Implements guardrails to prevent harmful outputs, detect prompt injections, and validate inputs/outputs.
• Content Moderation: Features include PII filtering, toxicity detection, and hallucination checking.
• Use Case: Integrate this Skill into a customer-facing chatbot to automatically filter out sensitive personal information, block inappropriate user requests, and ensure the LLM's responses are factual and non-toxic, thereby protecting users and maintaining brand integrity.

Quick Start

Install the nemoguardrails library by running 'pip install nemoguardrails'.