SkillRevise: 通过轨迹条件技能修订改进LLM撰写的智能体技能
SkillRevise: Improving LLM-Authored Agent Skills via Trace-Conditioned Skill Revision
AI总结：
提出SkillRevise框架，通过执行证据诊断、修复原则检索和执行锚定编辑，迭代优化初始技能，在SkillsBench上将基础智能体成功率从36.05%提升至61.63%，并展现跨模型迁移性。
摘要：
智能体技能是使LLM智能体能够执行工作流、验证约束并从故障中恢复的程序性工件。现有的自进化方法利用累积轨迹来优化技能，但在冷启动场景下（仅有一个初始的不完美技能可用）表现不佳。因此，技能构建默认采用专家编写或一次性LLM生成。专家编写的技能成本高昂，且可能与LLM智能体实际执行任务的方式不一致，而一次性生成的技能可能在语法上良好但在行为上薄弱。为弥合这一差距，我们提出SkillRevise，一个基于执行的框架，旨在迭代优化这些初始技能。SkillRevise从执行证据中诊断技能缺陷，从通用记忆中检索相关修复原则，并应用执行锚定编辑。通过重新执行候选技能并测量经验效用，它系统地保留最优技能版本。在三个基准测试和五个LLM上的评估表明，SkillRevise显著优于一次性基线，将SkillsBench上基础智能体的成功率从36.05%提升至61.63%。此外，修订后的技能展现出强大的跨模型迁移性，捕获了超越模型特定工件的通用程序性知识。
英文摘要：
Agent skills are procedural artifacts that enable LLM agents to execute workflows, verify constraints, and recover from failures. Existing self-evolving methods refine skills using accumulated trajectories. However, they struggle in cold-start settings, where only an initial, imperfect skill is available. Consequently, skill construction defaults to expert authoring or one-shot LLM generation. Expert-authored skills are costly and may not align with how LLM agents actually execute tasks, while one-shot generated skills can be syntactically well formed yet behaviorally weak. To bridge this gap, we propose SkillRevise, an execution-grounded framework designed to iteratively refine these initial skills. SkillRevise diagnoses skill defects from execution evidence, retrieves relevant repair principles from a general memory, and applies execution-anchored edits. By re-executing candidates and measuring empirical utility, it retains the best observed skill within the revision budget. Evaluated across three main benchmarks, two domain-specific studies, and six LLMs, SkillRevise substantially outperforms one-shot baselines, improving the base agent's success rate on SkillsBench from 36.05% to 61.63%. Furthermore, the revised skills transfer across both executors and task environments, suggesting that SkillRevise captures reusable procedural knowledge beyond any single executor. Our code is available at https://github.com/HKUST-KnowComp/skillrevise.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
