arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2605.11047cs.CRcs.AI
收藏

红队代理执行上下文:OpenClaw上的开放世界安全评估

Red-Teaming Agent Execution Contexts: Open-World Security Evaluation on OpenClaw

National University of Singapore(新加坡国立大学)

Hongwei Yao, Yiming Liu, Yiling He, Bingrun Yang

AI总结:

本文提出DeepTrap框架,通过黑盒轨迹优化发现OpenClaw中的上下文漏洞,展示上下文妥协可引发安全风险,强调需执行中心的安全评估。

摘要:

代理语言模型系统越来越多地依赖可变的执行上下文,包括文件、内存、工具、技能和辅助制品,从而产生超出显式用户提示的安全部署风险。本文提出了DeepTrap,一个自动框架,用于发现OpenClaw中的上下文漏洞。DeepTrap将对抗性上下文操纵建模为黑盒轨迹级优化问题,平衡风险实现、良性任务保留和隐蔽性。它结合了风险条件评估、多目标轨迹评分、奖励引导的束搜索和基于反射的深度探测,以识别高价值的受侵上下文。我们构建了一个包含42个案例的基准,涵盖六类漏洞和七个操作场景,并使用攻击和效用评分评估了九个目标模型。结果表明,上下文妥协可以诱导显著的不安全行为,同时保持用户面向任务的完成,证明最终响应评估是不足的。研究结果强调了对代理AI系统执行中心安全评估的必要性。我们的代码已发布在:https://github.com/ZJUICSR/DeepTrap

英文摘要:

Agentic language-model systems increasingly rely on mutable execution contexts, including files, memory, tools, skills, and auxiliary artifacts, creating security risks beyond explicit user prompts. This paper presents DeepTrap, an automated framework for discovering contextual vulnerabilities in OpenClaw. DeepTrap formulates adversarial context manipulation as a black-box trajectory-level optimization problem that balances risk realization, benign-task preservation, and stealth. It combines risk-conditioned evaluation, multi-objective trajectory scoring, reward-guided beam search, and reflection-based deep probing to identify high-value compromised contexts. We construct a 42-case benchmark spanning six vulnerability classes and seven operational scenarios, and evaluate nine target models using attack and utility grading scores. Results show that contextual compromise can induce substantial unsafe behavior while preserving user-facing task completion, demonstrating that final-response evaluation is insufficient. The findings highlight the need for execution-centric security evaluation of agentic AI systems. Our code is released at: https://github.com/ZJUICSR/DeepTrap

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑