让大模型自主生成复杂技能包，提升多步骤任务处理能力。
CoEvoSkills: Self-Evolving Agent Skills via Co-Evolutionary Verification
- 用协同进化机制自动生成多文件技能包，无需人工标注。
- 在SkillsBench上超越5个基线，在6种大模型上表现良好。
- 适合研究智能体自主能力与复杂任务处理的学者和开发者。
Anthropic提出将技能作为大模型智能体解决多步骤专业任务的核心能力，区别于单一工具调用，技能是包含多个互相关联文件的结构化集合。当前技能生成依赖人工标注，且存在人机认知错位问题，导致代理性能下降，已在SkillsBench评测中得到验证。为此，本文提出CoEvoSkills框架，使智能体能自主构建复杂多文件技能包。该框架通过一个技能生成器与一个代理验证器协同进化，后者在无真实测试内容的情况下提供有效反馈。在SkillsBench上，CoEvoSkills在Claude Code和Codex上均优于五个基线，并在六种其他大模型上表现出强泛化能力。代码已开源：https://github.com/Zhang-Henry/CoEvoSkills。
原文摘要 · Abstract (English)
Anthropic proposes the concept of skills for LLM agents to tackle multi-step professional tasks that simple tool invocations cannot address. A tool is a single, self-contained function, whereas a skill is a structured bundle of interdependent multi-file artifacts. Currently, skill generation is not only label-intensive due to manual authoring, but also may suffer from human--machine cognitive misalignment, which can lead to degraded agent performance, as evidenced by evaluations on SkillsBench. Therefore, we aim to enable agents to autonomously generate skills. However, existing self-evolving methods designed for tools cannot be directly applied to skills due to their increased complexity. To address these issues, we propose CoEvoSkills, a self-evolving skills framework that enables agents to autonomously construct complex, multi-file skill packages. Specifically, CoEvoSkills couples a Skill Generator that iteratively refines skills with a Surrogate Verifier that co-evolves to provide informative and actionable feedback without access to ground-truth test content. On SkillsBench, CoEvoSkills outperforms five baselines on both Claude Code and Codex, and generalizes strongly to six additional LLMs. The code is publicly available at https://github.com/Zhang-Henry/CoEvoSkills.
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。