Azure 架构指南：如何选择技能或子代理
原标题：Azure and Community Guidelines on Choosing Between a Skill or a Sub-Agent
AI 摘要
Azure 首席工程师 Kishorekumar Pattabiraman 在 Azure 架构博客中提出了在构建 AI 系统时选择技能（skill）或子代理（sub-agent）的实用标准，强调可重用性、简单性和长期可维护性。他指出了四个关键维度：迭代模型、语音保真度、人工门控位置和任务重复频率，并讨论了社区中关于两者权衡的讨论。文章还提到，在实践中技能和子代理可以组合使用，形成分层设计。
正文节选
In a recent Azure Architecture blog article, Azure lead engineer Kishorekumar Pattabiraman outlines practical criteria for choosing between skills, sub-agents, and other approaches when building AI systems, emphasizing reusability, simplicity, and long-term maintainability. According to Pattabiraman, teams often begin with the wrong question, focusing on which is the right model to use. The "first real fork" though, is not the model, but the architecture: Are you building a skill or a sub-agent?