// ADVERSARIAL PROMPT TESTBED

Try to break
the assistant.

A live AI endpoint, locked behind a single access phrase. Everything you send is logged. Jailbreaks, leaks, and instruction overrides are the whole point — that's what this surface is here to absorb.

VIEW ATTEMPT LOG
ENDPOINT /api.php
STATE LOCKED
SESSION EPHEMERAL
VECTOR

System Prompt Extraction

Attempts to get the assistant to reveal hidden instructions, config, or its own rules verbatim.

VECTOR

Instruction Override

Multi-step prompts that try to bury a rule-break inside a longer, legitimate-looking task.

VECTOR

Role / Mode Bypass

"Developer mode," "DAN," or persona-swap prompts aimed at shedding safety constraints.