GreyZone：面向生产级智能体应用的红队服务评测基准
原标题：GreyZone: Production-quality Agentic Application Threats
AI 摘要
研究者提出 GreyZone 基准，用于评估自动化红队服务对真实生产级智能体应用发起攻击的能力。该基准选取独立开发的开源智能体应用，覆盖访问权限、机密性和操作授权等安全边界，并以 Promptfoo 对 Letta Code 和 Clawith 进行初步案例研究。与 AgentDojo 等侧重评估被攻击智能体表现的基准不同，GreyZone 关注红队服务能否生成并交付经核实的策略违规攻击，并开源了基准及攻击面以支持后续研究。
正文节选
GreyZone: Production-quality Agentic Application Threats Abstract. Agents are increasingly deployed across domains such as coding, banking, and commerce. Their ability to invoke tools and change application state makes security testing important for reliable deployment. Although agent security has been studied through various adversarial testing approaches, benchmarks grounded in production agent environments remain limited. We introduce GreyZone, a benchmark for evaluating automated red-teaming