SHE: Trajectory-driven Safety Harness Evolution for LLM Agents

By Wanying Qu · Paper · cs.AI

The safety of large language model (LLM) agents depends not only on model weights but also on the agent harness that manages context, memory, tools, permissions, and runtime control. Existing safety mechanisms often treat the harness as a fixed deployment artifact, limiting their

Cs.ai

View original

HomeResourceLoading…