HarnessRisk is a lifecycle-oriented benchmark for evaluating safety in agent harnesses that manage tools, extensions, state, permissions, and external actions. It defines six operational phases—Harness Configuration, Capability Extension, Runtime Operation, State Persistence, Action Control, and Incident Recovery—and includes 128 sandboxed cases pairing benign user objectives with adversarial instructions. Across three harnesses, six language models, and 14 configurations, attack success rates vary from 12.6% to 80.9%, with the most vulnerable phase being Harness Configuration.
"whyItMatters":"The benchmark demonstrates that safety failures can arise in multiple harness responsibilities and that even explicit risk detection does not guarantee safe action, underscoring the need for comprehensive evaluation across model and harness configurations."
By Yajing Bai, Jinhao Duan, Jie Peng, Xianfeng Wu, Sijia Liu, Song Wang, Tianlong Chen
arXiv:2606. 18356v1 Announce Type: cross Abstract: Tool-using language-model agents introduce security failures that go beyond unsafe text: they can disclose protected objects, write persistent memory, send messages, modify databases, or trigger harmful code and tool effects.
By Yuchuan Tian, Mengyu Zheng, Haocheng Mei, Ye Yuan, Chao Xu, Xinghao Chen, Hanting Chen, Yu Wang
LLM-based agents leverage third-party skills to extend their capabilities in open-world scenarios. However, third-party skills can introduce extra security vulnerabilities, as seemingly harmless skills can contain latent safety risks that only emerge during actual execution.
arXiv:2609.39065v1 Announce Type: cross
Abstract: LLM agents increasingly rely on installable skills, which are packages of instructions, code, and resources that equip them with task-specific capabi...
By Yan Wang, Zhihao Zhang, Ke Chen, Kai Chen, Yaqin Zhang, Duohe Ma, Jun Dai, Xiaoyan Sun
arXiv:2610.03153v1 Announce Type: cross
Abstract: Workspace agents combine large language models with execution harnesses to perform stateful, multi-step tasks that access or modify external resource...
By Shiyi Kuang, Xuemei Luo, Kun Liu, Junhai Li, Rui Tian, Feng Shi, Bo Shen, Nianyu Li, Dehui Li, Ping Chen
The paper introduces skill cascading attacks, where a malicious goal is spread across multiple seemingly benign skills, causing harmful outcomes when combined. It presents SkillCascade, an automated red‑teaming framework, and releases SkillCascade‑Bench, a benchmark of 213 validated cascading test cases across various agent systems and domains. Experiments show that these cascaded interactions reliably induce harmful behaviors while evading existing per‑skill scanners and runtime monitors, revealing a gap between component‑level integrity and system‑level safety.
By Zihao Zhu, Siwei Lyu, Adel Bibi, Baoyuan Wu
The paper introduces PERSIST, a method that allows attackers to inject malicious actions into large language model agents by linking attacker-chosen actions to useful functions while minimizing source inspection and suspicion. PERSIST builds helper functions from simulated executions and embeds them in separate modules, enabling the agent to perform both user tasks and attacker objectives. Experiments on SkillsBench and Skill-Inject demonstrate that frozen skill packages transfer across multiple agent–model configurations, generalize to unseen tasks, and increase harmful-effect rates compared to baselines.
By Haochang Hao, Dehai Min, Zhifang Zhang, Yunbei Zhang, Miao Xu, Yingqiang Ge, Philip S. Yu, Lu Cheng
TRUSS is a framework that generates and verifies automated agent skills, ensuring they are both functionally effective and safe. It evaluates candidate skills against source evidence and nine safety properties, then tests them in a controlled environment to capture execution traces and identify failures. The system iteratively refines skills based on these results, achieving high precision in vulnerability detection and significantly improving task performance and security rates.
By Zhibo Zhang, Zhen Ouyang, Ling Shi, Kailong Wang
arXiv:2607. 20982v1 Announce Type: new Abstract: As large language model agents increasingly operate autonomously with access to tools and external environments, ensuring their safe and reliable behavior becomes critical.
By Vishal Ishwar Naik, Chenyu Xu, Donna Dong, Hussein Hassan, Abhishek Pradhan, Ofer Mendelevitch, Tallat Shafat, Humayun Irshad
arXiv:2607. 29254v1 Announce Type: new Abstract: AI agents extend large language models (LLMs) with external tools, enabling them to perform complex tasks and translate model outputs into consequential real-world actions.
By Minghui Pan, Jiayuxuan Yang, Yuanyuan Yuan, Yu Jiang, Zhenpeng Chen
arXiv:2608. 02683v1 Announce Type: cross Abstract: Large Language Model (LLM) agents rely on multi-stage agentic workflows, with stages such as memory, planning, and tool execution, to accomplish complex tasks.
By Zibo Xiao, Haoyu Wang, Jun Sun
TRUSS is a framework that generates and verifies automated agent skills, ensuring they are both functionally effective and safe. It first checks functional claims against evidence and evaluates artifacts against nine safety properties, then tests admitted skills in a controlled environment to capture execution traces and identify failures. The approach achieves perfect precision and recall in vulnerability detection, significantly reduces attack success rates, and boosts task effectiveness and security rates in skill generation benchmarks.