Showing cs.CRShow all
2 papers · 1 filter
cs.CR2026
Beyond Direct Access: Resource Hijacking in LLM Agents
Puyu Zeng, Qibing Ren, Mingang Chen +1
Large language model agents are increasingly connected to high-value resources, including external APIs, GPUs and servers, and workflows such as deployment and approval. Existing a…
cs.CR2026
ColluSkill: Adversarial Cross-Skill Composition for Evading Agent Skill Scanners
Puyu Zeng, Simeng Qin, Jingzhi Li +3
Agent skills are emerging as an important attack surface in LLM-based agent systems. Through an empirical study of existing skill scanners, we find that current defenses mainly ins…