HazardAuditor: From Executable Threats to Safer Computer-Use Agents
Computer-use agents increasingly interact with browsers, terminals, file systems, and external services, introducing safety risks that emerge through runtime behavior rather than generated content alone. Existing guard models target static prompts and responses and are poorly suited to agent execution; existing executable safety platforms produce evaluation verdicts rather than
Reading in Plain English mode — toggle below to switch
Computer-use agents increasingly interact with browsers, terminals, file systems, and external services, introducing safety risks that emerge through runtime behavior rather than generated content alone. Existing guard models target static prompts and responses and are poorly suited to agent execution; existing executable safety platforms produce evaluation verdicts rather than
As AI agents become more autonomous and interact with real-world systems, ensuring their safety and preventing unintended hazardous actions is paramount. This work provides a crucial framework for evaluating and supervising agent behavior across diverse platforms, addressing a critical gap in current safety methodologies.