AI exploit bank
AI exploit bank
the most comprehensive agentic exploit dataset -- a growing bank of attack probes the AI red team fires at your agents, pre-seeded across OWASP LLM Top-10 + MITRE ATLAS, submittable, and federated from the open AI-security databases.
About
The reusable dataset the AI red team (aitest) draws from. Ships pre-seeded with curated agentic probes -- prompt injection (direct and indirect/RAG), jailbreaks, system-prompt and PII/secret exfiltration with canary tokens, excessive agency and unsafe tool use, insecure output handling (XSS/markdown), RCE/SQLi/SSRF via agents, denial-of-wallet, training/memory poisoning, multi-agent confused-deputy, and harmful/misinformation -- each mapped to an OWASP LLM Top-10 risk and/or a MITRE ATLAS technique with a detector and an expected outcome. Takes new probes you submit in the same shape, and federates the open AI-security databases (garak, MITRE ATLAS, OWASP, promptfoo/PyRIT) so it receives the world's existing attacks over time. Every plan the red team freezes is a snapshot of this dataset.
Tags
People behind this thing
Publisher · 919925188036
Explore, share & get help
Free
Personal, professional, publisher, or enterprise
yours the moment you get it — nothing to install
Get itHow it works
- 1 · Get itOne tap records the grant and lands the class on your zero — nothing to install.
- 2 · It stays currentYour zero pulls the latest from the store over the air, in the background — no app to update, ever.
- 3 · Evolve itTell it what to change — it gets better by conversation, and the new version ships over the air too.