There are 3 attacks in the information-disclosure category. Each shows its technique id, mode, standards mapping, description, and the exact command to run it (swap the provider and target as needed).
​
API Key Extraction
critical
api-key-extraction · multi-turn · MITRE: AML.T0083, AML.T0056Multi-turn escalation from architecture questions to extracting live API keysRun it:
ai-blackteam run -p anthropic -a api-key-extraction -t "your target prompt"
​
Cross-Session Leak
high
cross-session-leak · multi-turn · MITRE: AML.T0056, AML.T0080.002Pretend to continue a previous conversation to extract data from other sessionsRun it:
ai-blackteam run -p anthropic -a cross-session-leak -t "your target prompt"
​
Prompt Leaking
high
prompt-leaking · multi-turn · OWASP: LLM07 · MITRE: AML.T0056, AML.T0069.002Indirect system prompt extraction via creative reformatting (poem, JSON, code)Run it:
ai-blackteam run -p anthropic -a prompt-leaking -t "your target prompt"