SyncAI.news, a Varaisys broadcasting
SkillAtlas: An Attack Trace Library for Agent Skills
YT

Yuxin Tian, Zenghao Duan, Liang Pang, Zhiyi Yin, Xueqi Cheng

· 1 min read

ResearcharXiv cs.AI

SkillAtlas: An Attack Trace Library for Agent Skills

arXiv:2609.13353v2 Announce Type: replace-cross Abstract: Agent skills are reusable units for language-model agents, but their risks emerge through model decisions, user context, tool calls, and execution feedback rather than through stable signatures or a single sandbox run. Existing static, dynamic, and benchmark-style evaluations rarely preserve public evidence that can be inspected, searched, and reused. We present SkillAtlas, a hosted attack trace library that converts private agent-skill security report bundles into reviewed, redacted, and searchable public cases. The library contains 3,014 cases, 6,589 traces, 151,131 steps, 233 affected skills, and 8 risk categories; 42.5% of successful cases first become successful after a non-success initial round, and trajectory-grounded labels improve pre-execution guard accuracy to 0.770.

Original source

This story was published by arXiv cs.AI and written by Yuxin Tian, Zenghao Duan, Liang Pang, Zhiyi Yin, Xueqi Cheng. SyncAI.news shows a preview; the complete article is on the publisher's site.

Read the full story on arxiv.org

Similar News