AI RESEARCH

RouteGuard: Internal-Signal Detection of Skill Poisoning in LLM Agents

arXiv CS.AI

ArXi:2604.22888v1 Announce Type: cross