SKILL
safety-alignment-nemo-guardrails
NVIDIA's runtime safety framework for LLM applications. Features jailbreak detection, input/output validation, fact-checking, hallucination detection, PII filtering, toxicity detection. Uses Colang 2.0 DSL for programmable rails. Production-ready, runs on T4 GPU.
Target: Learning & research. Automatically collected · unverified.
Use with Cursor
A guide for this combination is not available yet, and compatibility has not been verified.
Requirements
- Only the author’s Skill metadata is included. Read the full instructions to check dependencies, licensing and conditions of use.
Choose an AI tool with documented instructions, or save the resource and request a guide after the page loads.
Read the author’s documentation ↗