6 citations · 6 across the 1 of their papers we have counts for
Showing cs.CRShow all
2 papers · 1 filter
cs.CR2026
Towards Elastic Least Privilege for Agent Planning
Simon Parkinson, Zhipeng Wang, Mustafa Mustafa
General-purpose AI agents require authority that changes between tasks and during execution. Common deployments trade broad permissions that expose unnecessary authority and narrow…
cs.CR2026
Does Showing Agents the Privilege of Their Tools Reduce Over-Privileged Tool Selection?
Simon Parkinson, Mauro Vallati, Zhipeng Wang +1
When several tools can complete a task, an agent that picks the most powerful one holds more authority than the task needs. Recent work demonstrates that language-model agents do t…