EvoSafeHarness: Evolving Model- and Domain-Specific Harnesses for Securing Agents

Abstract

Large Language Model (LLM) agents are turning language into real-world effects, making safety necessary against both indirect prompt injections and direct harmful requests. System-level safety harnesses add an enforcement layer beyond model-level defenses, but existing harnesses are usually designed once by experts and applied across heterogeneous models and domains. Effective protection is deployment-dependent: models differ in how much enforcement they need before utility declines, while domai

Categories

Framework mappings

Suggested from the entry's categories.

@misc{li2026evosafeharness,
  title = {{EvoSafeHarness: Evolving Model- and Domain-Specific Harnesses for Securing Agents}},
  author = {Nanxi Li and Yingzi Ma and Yulong Cao and Edward Suh and Bo Li and Dawn Song and Chaowei Xiao},
  year = {2026},
  month = sep,
  eprint = {2609.05903},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2609.05903}
}