Every model has an attack surface.

Start here

The reference pages on this site, independent of publication date. Each one is sourced to published research and updated as the technique or the defense against it moves.

• Ten documented jailbreak patterns Each pattern with the mechanism, the reported success rate from its source, and the signal that detects it.
• Tool-call hijacking in agentic systems How injected context turns an agent's own function calls against it, where attackers intercept the tool loop, and the confirmation gates that stop it.
• Many-shot jailbreaking How hundreds of in-context demonstrations erode safety training, and why longer context windows make it worse.
• Model extraction via black-box queries Query budgets, extraction fidelity, and how each deployed countermeasure is defeated.
• Supply chain attacks on AI models Dataset poisoning, malicious fine-tuning, and serialization exploits that compromise a model before it ever reaches production.
• Why static jailbreak filters fail What happens to twelve published defenses when the attacker is allowed to adapt to them.
• Red Team Plan Builder Pick the components and trust boundaries a target has; get an ordered checklist of the techniques that reach it.