When Lower Privileges Suffice: Investigating Over-Privileged Tool Selection in LLM Agents
This work addresses the safety-critical problem of privilege escalation in LLM agents, which is underexplored and poses risks in autonomous tool use.
The paper studies over-privileged tool selection in LLM agents, where agents choose higher-privilege tools despite sufficient lower-privilege alternatives. Introducing ToolPrivBench, they find that such behavior is common across eight domains and five risk patterns, and is amplified by transient failures. They propose a privilege-aware post-training defense that reduces unnecessary high-privilege tool use while preserving capabilities.
As LLM agents increasingly select tools autonomously, their choices among tools with different privileges become safety-relevant. However, prior tool-selection studies focus on safety-agnostic metadata preferences, leaving privilege-sensitive choices underexplored. To address this gap, we study over-privileged tool selection, in which an agent selects or escalates to a higher-privilege tool despite a sufficient lower-privilege alternative. We introduce ToolPrivBench to evaluate whether agents choose higher-privilege tools despite sufficient lower-privilege alternatives, measuring both initial selection and escalation after transient tool failures. Across eight domains and five recurring risk patterns, we find that over-privileged tool selection is common among mainstream LLM agents and is further amplified by transient failures. We further find that general safety alignment does not reliably transfer to least-privilege tool choice, while prompt-level controls provide only limited mitigation under transient failures. We therefore introduce a privilege-aware post-training defense that teaches agents to prefer sufficient lower-privilege tools and escalate only when necessary. Our mitigation experiments show that this defense substantially reduces unnecessary high-privilege tool use while preserving general capabilities.