VishBox v2: A Multi-Agent System for Adaptive Voice Phishing Simulation

Sungmi Park, Daon Choi, YOONMO YANG, Hong Yunyi, Heedou Kim

Keywords: multi-agent system, MAS, voice-phishing, synthetic data, automated decision making

TL;DR: A multi-agent architecture that simulates adaptive, multi-round voice phishing conversations with procedural grounding and emotion-driven vulnerability tracking for safer security research and training.

Abstract: Voice phishing is a multi-round social engineering attack in which strategy and victim psychology co-evolve, yet real transcripts are rarely accessible for systematic analysis. We present VishBox v2, a multi-agent architecture that generates structured phishing simulations grounded in crime-script procedures and persuasion principles. A Main Agent orchestrates a Dialogue Agent and a Tactic Search Agent, combining multi-round dialogue generation, web-based tactic mining, and emotion-driven vulnerability tracking. Across 571 rounds, results including police-expert evaluation support procedural realism and show that VishBox v2 captures tactic concentration, vulnerability transitions, and web-search-induced procedural disruptions. The framework provides a controlled foundation for safer red-teaming and security training research.

Submission Type: Emerging

Submission Number: 517