好技能要留关键信息，删了可能让大模型更费钱。
What Should a Skill Remember? Quality--Cost Trade-offs in Cost-Aware Skill Rewriting for Language Model Agents
- 用保留关键信息的策略重写技能，避免过度压缩
- 实验显示不同任务适合不同锚点，无通用最优模板
- 新方法降低总成本7%，适合关注效率的开发者
大型语言模型代理越来越多依赖可复用的技能：编码工作流、工具使用、实现模式、验证检查和领域规则的流程文档。技能重写常被视为提示压缩，但更短的技能可能因移除稀疏的操作锚点而增加代理成本，这些锚点原本用于防止探索偏差、调试困难和恢复失败。本文从经济视角研究技能重写。在受控框架下，分析技能结构，采用信息保全策略重写技能，并在固定任务指令、环境和验证器条件下评估结果。基于SkillsBench的实验揭示不同策略存在明显的质量-成本权衡：API/代码锚定、工作流保护和规则/公式锚定分别适用于不同任务类型，无普遍最优模板。主测试集上，学习到的策略使总成本降低7.0%，下游代理令牌成本降低6.0%；在冻结跨模型迁移中，对应降幅平均达14.7%和13.7%，验证器性能保持不变。结果表明，技能设计应作为成本感知的操作知识工程，而非简单提示压缩。资源：https://github.com/1Reminding/Skill_EE。
原文摘要 · Abstract (English)
Large language model agents increasingly rely on skills: reusable procedural documents encoding workflows, tool use, implementation patterns, validation checks, and domain rules. Skill rewriting is often treated as prompt compression, but shorter skills can make agents more expensive by removing sparse operational anchors that prevent exploration, debugging, and recovery. We study skill rewriting through this economic lens. Our controlled framework profiles skill structure, rewrites skills using information-preservation strategies, and evaluates the rewrites under fixed task instructions, environments, and verifiers. Experiments on SkillsBench reveal distinct quality--cost trade-offs across strategies: API/code anchoring, workflow guarding, and rule/formula anchoring benefit different task families, with no universally dominant template. In the main held-out evaluation, the learned policy reduces total cost by 7.0% and downstream agent-token cost by 6.0%; in frozen cross-model transfer, the corresponding reductions average 14.7% and 13.7%, while verifier quality is preserved. These results position skill design as cost-aware operational knowledge engineering rather than prompt compression. Resources: https://github.com/1Reminding/Skill_EE.
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。