Garak
by NVIDIAOpen-source LLM vulnerability scanner for probing model behavior, prompt attack surfaces, and safety failures.
2026-04-27credited to NVIDIA
Contributor credited on accepted HeyClaude registry entries.
Open-source LLM vulnerability scanner for probing model behavior, prompt attack surfaces, and safety failures.
Official NVIDIA-verified Agent Skills catalog for AI agents working with CUDA-X, RAG Blueprint, NemoClaw and OpenClaw, NeMo, TAO, cuOpt, Omniverse, Physical AI, vision AI, simulation, robotics, and GPU workflows.
Apache-2.0 security scanner from NVIDIA for AI agent skills, with static pattern checks, optional LLM semantic analysis, MCP least-privilege and tool poisoning analyzers, OSV.dev vulnerability lookups, risk scoring, and terminal, JSON, Markdown, and SARIF reports.
No reviewed entries yet.