가이드 연계 퀴즈 · 하드 레벨

The Dual-LLM Pattern Against Prompt Injection Quiz

Tests trust boundaries, limited extraction, validation, tool authorization, and the limits of dual-model defenses against prompt injection.

관련 가이드 경로Dual LLM Pattern Against Prompt Injection
질문 1 의 8

Which model is permitted to execute an approved tool action in the pattern described?