SkillOpt: 自我进化智能体技能的执行策略
SkillOpt: Executive Strategy for Self-Evolving Agent Skills
AI总结：
提出SkillOpt，一种系统性的可控文本空间优化器，通过分离的优化器模型对技能文档进行有界编辑，并仅在严格改善验证分数时接受编辑，从而稳定训练技能，在六个基准测试中全面优于现有方法。
摘要：
当前的智能体技能要么是手工制作的，要么是一次性生成的，要么通过松散控制的自我修订来进化，这些方法都不像深度学习优化器那样作用于技能，并且都无法在反馈下可靠地改进其起点。我们认为，技能应该作为冻结智能体的外部状态进行训练，并遵循使权重空间优化可复现的相同原则。据我们所知，SkillOpt是第一个系统性的可控文本空间优化器，用于智能体技能：一个独立的优化器模型将带分数的轨迹转换为对单个技能文档的有界添加/删除/替换编辑，并且仅当编辑严格改善保留验证分数时才接受编辑。文本学习率预算、拒绝编辑缓冲区和逐轮慢/元更新使得技能训练稳定，同时在部署时无需增加推理时的模型调用。在六个基准测试、七个目标模型和三个执行框架（直接对话、Codex、Claude Code）中，SkillOpt在所有52个评估的（模型、基准、框架）单元上取得最佳或并列最佳，并击败了每个单元上的所有竞争者，包括人类、一次性LLM、Trace2Skill、TextGrad、GEPA和EvoSkill技能。在GPT-5.5上，它在直接对话中将平均无技能准确率提高了23.5个百分点，在Codex智能体循环中提高了24.8个百分点，在Claude Code中提高了19.1个百分点。迁移实验进一步表明，优化后的技能工件在跨模型规模、在Codex和Claude Code执行环境之间迁移以及迁移到邻近的数学基准测试时，无需进一步优化即可保留其价值。代码：https://aka.ms/skillopt
英文摘要：
Agent skills today are hand-crafted, generated one-shot, or evolved through loosely controlled self-revision, none of which behaves like a deep-learning optimizer for the skill, and none of which reliably improves over its starting point under feedback. We argue the skill should instead be trained as the external state of a frozen agent, with the same discipline that makes weight-space optimization reproducible. SkillOpt is, to our knowledge, the first systematic controllable text-space optimizer for agent skills: a separate optimizer model turns scored rollouts into bounded add/delete/replace edits on a single skill document, and an edit is accepted only when it strictly improves a held-out validation score. A textual learning-rate budget, rejected-edit buffer, and epoch-wise slow/meta update make skill training stable while adding zero inference-time model calls at deployment. Across six benchmarks, seven target models, and three execution harnesses (direct chat, Codex, Claude Code), SkillOpt is best or tied on all 52 evaluated (model, benchmark, harness) cells and beats every per-cell competitor among human, one-shot LLM, Trace2Skill, TextGrad, GEPA, and EvoSkill skills. On GPT-5.5 it lifts the average no-skill accuracy by +23.5 points in direct chat, by +24.8 inside the Codex agentic loop, and by +19.1 inside Claude Code. Transfer experiments further show that optimized skill artifacts retain value when moved across model scales, between Codex and Claude Code execution environments, and to a nearby math benchmark without further optimization. Code: https://aka.ms/skillopt
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。
