OWASP LLM Top 10
-
AI Red Teaming Tools for LLM Safety: 6 Options Compared
Promptfoo, garak, PyRIT, DeepTeam, Microsoft Foundry and Prisma AIRS compared on attack coverage, multi-turn depth, CI fit and agent-testing gaps.
-
What Is an LLM Firewall? Prompt-Layer Defenses Explained
An LLM firewall inspects prompts and completions for injection, PII, and unsafe content. What these controls catch, where they sit, and what they miss.
-
How to Detect Jailbreak Prompts: A Practitioner's Guide
How to detect jailbreak prompts with perplexity heuristics, trained classifiers, and multi-turn scoring, plus the blind spots of each approach.
-
OWASP Prompt Injection (LLM01): How to Mitigate Attacks
No single control stops prompt injection. This guide covers the layered defenses that reduce risk in production, from input hardening to monitoring.