Doppelganger Method: Breaking Role Consistency in LLM Agent via Prompt-based Transferable Adversarial Attack

원문 초록

Since the advent of large language models, prompt engineering now enables the rapid, low-effort creation of diverse autonomous agents that are already in widespread use. Yet this convenience raises urgent concerns about the safety, robustness, and behavioral consistency of the underlying prompts, along with the pressing challenge of preventing those prompts from being exposed to user's attempts. In this paper, we propose the ''Doppelganger method'' to demonstrate the risk of an agent being hijacked, thereby exposing system instructions and internal information. Next, we define the ''Prompt Alignment Collapse under Adversarial Transfer (PACAT)'' level to evaluate the vulnerability to this adversarial transfer attack. We also propose a ''Caution for Adversarial Transfer (CAT)'' prompt to counter the Doppelganger method. The experimental results demonstrate that the Doppelganger method can compromise the agent's consistency and expose its internal information. In contrast, CAT prompts enable effective defense against this adversarial attack.