Results for “forward-secrecy”

Test LLM applications for system prompt leakage using manual payloads, garak, and Promptfoo to extract embedded secrets and routing logic.

Simulate an insider threat or post-breach attacker to identify lateral movement paths, privilege escalation vectors, and sensitive data exposure within a corporate network.