AI-SYS-PROMPT-LEAKDiscovery

System Prompt Extraction

Coax the model into echoing its hidden system prompt via role-play / repetition / encoding tricks — reveals business logic, allowlists, secrets.

§ Where this technique fits

AI-SYS-PROMPT-LEAK is catalogued under the Discovery tactic of the offensive-security kill-chain. It appears in 1 approved dossier in the registry, typically at step 2 on average.

§ Dossiers chaining this technique

Direct prompt injection → exfil another user's data
Multi-tenant LLM assistant. Attacker's prompt overrides instructions and tricks the model into emitting another user's session content / RAG-cached data.
step 2 / 5

§ What commonly comes next

01
Direct Prompt Injection
AI-PROMPT-INJECT · Initial Access
seen 1×

§ Where this technique fits

§ Dossiers chaining this technique

Direct prompt injection → exfil another user's data

§ What commonly comes next