paperAugust 2026Unreviewed

SkillWatermark: An Embedded Skill Watermark of Progressive Privacy Inference via Benign Prompts

Yu Li, Liqi Zhuang, Dong Wei, Jiwen Luo, Hang Zhang, Meng Zhang, Xiaona Li, Weiqing Huang

Abstract

Skills for large language model (LLM) agents have been widely deployed across diverse application domains. However, we observe that these skills generate specific traffic patterns during execution. In this paper, we design a pipeline that generates specific traffic patterns by inserting carefully designed skill descriptions, which we term skill watermarks, so that a passive network attacker can establish a covert channel to encode private information within observable traffic across multiple con

Cite

@misc{li2026skillwatermark,
  title = {{SkillWatermark: An Embedded Skill Watermark of Progressive Privacy Inference via Benign Prompts}},
  author = {Yu Li and Liqi Zhuang and Dong Wei and Jiwen Luo and Hang Zhang and Meng Zhang and Xiaona Li and Weiqing Huang},
  year = {2026},
  month = aug,
  eprint = {2608.16026},
  archivePrefix = {arXiv},
  url = {https://www.semanticscholar.org/paper/ae12dba64123005c4bae0df7c084a9eadc318a88}
}