Awesome AI Security Tools
Section: Scanners, Evals & Guardrails · 🟢🔬 — Python framework for adversarial attacks, data augmentation, and training for NLP models; useful for robustness testing beyond chat-only LLM scanners. · updated 2026-08-15)
Entry
Appears in 6 awesome lists
Python framework for adversarial attacks, data augmentation, and model training in NLP. Augment datasets to increase model robustness and generate adversarial examples. MIT licensed.
Section: Scanners, Evals & Guardrails · 🟢🔬 — Python framework for adversarial attacks, data augmentation, and training for NLP models; useful for robustness testing beyond chat-only LLM scanners. · updated 2026-08-15)
Section: Tools · A Python framework for adversarial attacks, data augmentation, and model training in NLP.
Section: Adversarial Attack · (from UVa) - A Python framework for adversarial attacks, data augmentation, and model training in NLP.
Section: Libraries · Adversarial attacks, adversarial training, and data augmentation in NLP
Section: 1. Core Frameworks & Libraries · Python framework for adversarial attacks, data augmentation, and model training in NLP. Augment datasets to increase model robustness and generate adversarial examples. MIT licensed.
Section: LLM Security & AI Security · Framework for adversarial attacks in NLP.
🟢 — LLM eval + red-teaming/pentesting CLI with 50+ attack plugins (MIT). Note: OpenAI announced an acquisition agreement in March 2026; remains MIT-licensed — track governance. · updated 2026-10-13); Related: Promptfoo CI Action (checkout evaluation, not base/head diff; defaults to latest,…
🟢 — Python framework for adding input/output guards, validators, structured-output controls, and Guardrails Hub checks to LLM applications. (Guardrails AI) · updated 2026-08-26); Related: NeMo Guardrails · LLM Guard · Pydantic AI Shields integration (README recommends a successor)
NVIDIA's programmable guardrails toolkit: define input, dialog, retrieval, execution, and output rails that intercept the agent loop at five distinct layers using the Colang DSL. The execution rail layer specifically governs what tools the LLM can invoke and what their inputs/outputs may contain —…
🙌 OpenHands: Code Less, Make More. (formerly OpenDevin), a platform for software development agents powered by AI. github
Stanford NLP Python library for 100+ human languages. State-of-the-art neural pipelines for tokenization, NER, parsing, and sentiment analysis with pre-trained models. Apache 2.0 licensed.