A public jailbreak archive
L1B3RT4S
Adversarial prompts and techniques, collected in the open. Material to study, reproduce, challenge, and take further.
VII · The Wings
Go where the questions lead.
Adversarial inquiry, experimental AI, and a closer look at what lies beneath the surface.
System prompt extraction & transparency
CL4R1T4S
A public archive of extracted system prompts, tool definitions, and behavioral guidelines from AI models and agents. Studying the instructions beneath the interface—and bringing them into the open.
BT6 · Black Hat USA 2026
Kinetic Prompt Injection
Agent compromise with a physical blast radius. A BT6 presentation on what happens when prompt injection reaches systems that act in the world.
The ghost in the weights
Poison in the Pipeline
An investigation into training-data poisoning and model behavior, through the Basilisk Venom experiment.
Experimental xenolinguistics
GLOSSOPETRAE
Forging new languages for artificial minds. Open experiments in language learning, stealth communication, and the blind spots of AI monitors.
Join the mission
Bring the strange question.
Study model behavior, language, agents, and the boundaries we haven’t mapped yet.
Join Research