Get the latest tech news
Show HN: A minimal TS library that generates prompt injection attacks
Enterprise-grade prompt injection testing for AI security professionals
A minimal TypeScript library with 25+ curated prompt injection patterns from leading security research. Jailbreaking (5 patterns) Role-play and persona-based attacks that attempt to bypass AI safety guidelines through character roleplay and fictional scenarios. Logic Traps (6 patterns) Sophisticated reasoning exploits using hypothetical scenarios, false urgency, and academic authority to manipulate responses.
Or read this on Hacker News