1
STAC: When Innocent Tools Form Dangerous Chains for LLM Agents
揭秘看似无害的工具如何被组装成攻击链,威胁LLM Agent安全,一篇值得关注的漏洞挖掘研究。
arXiv:2509.25624v3 Announce Type: replace-cross Abstract: As LLMs advance into autonomous agents with tool-use capabilities, they introduce security c…