paperMay 2026Unreviewed

When Safe Skills Collide: Measuring Compositional Risk in Agent Skill Ecosystems

Su Wang, Pin Qian, Yihang Chen, Junxian You, Xiaoyuan Wang, Xiaochong Jiang, Lifei Liu, Haoran Yu, Jingzhou Xu

Abstract

LLM agents increasingly rely on community-contributed skills that expand an agent's operational capability set. We study a core safety problem in agentic AI systems: whether individually safe skills can compose into unsafe installed skill sets. We present SkillReact, a compositional security measurement framework with three components: a deterministic static-composition benchmark, a two-rater LLM-assisted human-adjudication pipeline, and an action-based exploitability harness. On 1,520 ClawHub s

Cite

@misc{wang2026when,
  title = {{When Safe Skills Collide: Measuring Compositional Risk in Agent Skill Ecosystems}},
  author = {Su Wang and Pin Qian and Yihang Chen and Junxian You and Xiaoyuan Wang and Xiaochong Jiang and Lifei Liu and Haoran Yu and Jingzhou Xu},
  year = {2026},
  month = may,
  eprint = {2606.00448},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2606.00448}
}