Awesome AI Security Tools
Section: Scanners, Evals & Guardrails · 🟢 — Programmable guardrails (input/output/dialog/retrieval rails) for LLM apps. (NVIDIA) · updated 2026-10-10); Related: Backtranslation-defense reproduction (historical; legacy FastChat/GPU and model judges, separate bundled-code terms) · Jailbreak-defense research library (historical; local…