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

ArXiv CS.AI2026-06-16NEWSen作者: Hongwei Yao, Yiming Liu, Yiling He, Bingrun Yang

详细信息

来源站点
ArXiv CS.AI
作者
Hongwei Yao, Yiming Liu, Yiling He, Bingrun Yang
文章类型
NEWS
语言
en
发布日期
2026-06-16

摘要

arXiv:2605.11047v2 Announce Type: replace-cross Abstract: 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.

相关事件

暂无数据

相关公司

暂无数据

相关人物

暂无数据

相关技术

暂无数据