构建技能依赖图，让大模型智能选技能并省下大量算力。
Graph-of-Skills: Dependency-Aware Structural Retrieval for Massive Agent Skills
- 离线建技能依赖图，推理时按依赖关系精准选技能。
- 在多场景测试中，奖励提升最高25.55%，token消耗减少56.72%。
- 适合需要高效调用成百上千技能的复杂智能体系统。
现代大模型智能体越来越依赖可复用的技能，随着与个人应用、浏览器等交互增多，技能库可能扩展至数千个。规模扩大带来两大挑战：一是加载全部技能会填满上下文窗口，导致成本高、幻觉多、延迟大；二是语义检索只找主题相关技能，忽略上下游依赖关系，导致选中的技能包无法执行。本文提出Graph-of-Skills（GoS），一种推理时的结构化检索层。GoS离线构建可执行的技能图，推理时通过混合语义-词法种子、反向感知个性化页面排名和上下文预算填充，检索出受限且依赖感知的技能包。在SkillsBench和ALFWorld上，GoS在三种模型（Claude Sonnet 4.5、MiniMax M2.7、GPT-5.2 Codex）上均显著提升奖励并节省token。在SkillsBench上，使用GPT-5.2 Codex时，奖励峰值提升25.55%，总token减少56.72%。消融实验验证该效果在200到2000个技能的库中均成立。
原文摘要 · Abstract (English)
Modern LLM agents increasingly rely on reusable skills, and as they interact with personal applications, web browsers, and other interfaces, skill libraries can scale to thousands of skills. Scaling to larger skill sets introduces two key challenges. First, loading the full skill set saturates the context window, driving up token costs, hallucination, and latency. Second, semantic retrieval surfaces topically relevant skills but misses their prerequisite chain of upstream and downstream skills, creating a prerequisite gap that leaves the retrieved bundle execution-incomplete. In this paper, we present Graph-of-Skills (GoS), an inference-time structural retrieval layer for large skill libraries. GoS constructs an executable skill graph offline from skill packages, then at inference time retrieves a bounded, dependency-aware skill bundle through hybrid semantic-lexical seeding, reverse-aware Personalized PageRank, and context-budgeted hydration. On SkillsBench and ALFWorld, GoS consistently delivers substantial reward improvements and token savings across three model families (Claude Sonnet 4.5, MiniMax M2.7, and GPT-5.2 Codex). On SkillsBench, GoS achieves a peak reward increase of 25.55% while reducing total tokens by 56.72% over the vanilla full skill-loading baseline using GPT-5.2 Codex. Ablations confirm this pattern across skill libraries from 200 to 2,000 skills.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。