arXiv cs.CL· Wenkai Yu, Ning Ding, Wentao Zhang, Yehui Tang·· 11 小时前AI 评分38
从流程到权限：从 Agent Skills 学习精确的子智能体授权
From Procedures to Permissions: Learning Precise Subagent Authorization from Agent Skills
AI 导读
研究提出利用 Agent Skills 中的流程知识作为监督信号，为子智能体学习精确的工具与工作区权限授权。在 ClawArena-Team 上，该方法使 Qwen3.5-35B-A3B 的工具权限精度提升 13.4 个百分点、工作区权限精度提升 24.1 个百分点，整体任务成功率提升 11.6 个百分点。
正文 · 原文
Abstract:As language agents tackle increasingly complex tasks, they increasingly rely on multiple agents to coordinate specialized work, requiring decisions not only about what each subagent should do but also which tools and workspace resources it should access. Existing agent systems provide mechanisms for controlling tools and execution environments, but do not directly specify the resources required by each delegated subtask. Moreover, task outcomes and execution trajectories alone provide limited supervision for determining whether a subagent has received appropriate permissions. We observe that Agent Skills encode procedural knowledge describing the operations, tools, and files required at each step, providing a natural source of supervision for authorization. Building on this observation, we synthesize tasks and step-by-step procedures from Skills, execute them through subagent delegation in interactive environments, and use verified trajectories to construct training data for task execution and authorization. On ClawArena-Team, our method improves tool permission precision by 13.4 percentage points and workspace permission precision by 24.1 percentage points, while increasing overall task success by 11.6 percentage points for Qwen3.5-35B-A3B. On Terminal-Bench 2.1 and SkillsBench, our method remains competitive with existing agent data synthesis approaches. These results show that the procedural knowledge embedded in Agent Skills provides a source of supervision for precise subagent orchestration that task outcomes alone cannot provide.
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2611.08905 [cs.CL] |
| (or arXiv:2611.08905v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2611.08905 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Wenkai Yu [view email]
[v1]
Fri, 6 Nov 2026 02:46:06 UTC (970 KB)
来源：arXiv cs.CL · arxiv.org