# Claude Code Skills 模块构建与AI编程辅助指南 ## Claude Code Skills 提示词模块化构建方法 **什么是Claude Code Skills?** Claude Code Skills(智能体技能)是由Anthropic推出的一种"**技能包**"机制,用于封装专业知识和工作流程,使大模型(如Claude)能够按需加载相应模块,完成特定领域的任务[\[1\]](https://ide.unitmesh.cc/spec/claude-skill#:~:text=%E4%BB%80%E4%B9%88%E6%98%AF%20Claude%20Skills%EF%BC%9F)[\[2\]](https://www.bilibili.com/opus/1132124852115734530#:~:text=Claude%20Skills%20%E6%9C%AC%E8%B4%A8%E4%B8%8A%E6%98%AF%E4%B8%80%E7%A7%8D,)。每个Skill本质上是一个**自包含的功能模块**,包含了**元数据、指令和资源**等要素。与传统一次性的大段 Prompt 不同,Skills采用**渐进式披露**架构,将上下文按需逐级加载,提高了上下文利用效率[\[3\]](https://claudecn.com/#:~:text=Agent%20Skills%20%E5%8F%AF%E5%A4%8D%E7%94%A8%E7%9A%84%E6%8A%80%E8%83%BD%E7%B3%BB%E7%BB%9F%EF%BC%8C%E8%AE%A9%20Claude%20%E6%8E%8C%E6%8F%A1%E7%89%B9%E5%AE%9A%E9%A2%86%E5%9F%9F%E4%B8%93%E4%B8%9A%E7%9F%A5%E8%AF%86%E3%80%82%E5%8C%85%E6%8B%AC,%E8%BF%9E%E6%8E%A5%E5%A4%96%E9%83%A8%E5%B7%A5%E5%85%B7%E5%92%8C%E6%95%B0%E6%8D%AE%E6%BA%90%EF%BC%8C%E6%89%A9%E5%B1%95%20Claude%20%E7%9A%84%E8%83%BD%E5%8A%9B%E8%BE%B9%E7%95%8C%EF%BC%8C%E6%9E%84%E5%BB%BA%E5%BC%BA%E5%A4%A7%E7%9A%84%20AI%20%E5%BA%94%E7%94%A8%E7%94%9F%E6%80%81%E3%80%82)[\[4\]](https://claudecn.com/blog/claude-skills-architecture/#:~:text=,3.2.2%20Markdown%20%E6%8C%87%E4%BB%A4%E6%AD%A3%E6%96%87)。换言之,我们可以把Claude Skills理解为AI工具箱中的插件,将反复使用的提示语和脚本封装成可复用模块,以便在需要时自动调用,从而避免每次手动重复提示[\[5\]](https://cloud.tencent.com/developer/article/2616585#:~:text=Claude%20Skills%E6%98%AFAnthropic%E6%8E%A8%E5%87%BA%E7%9A%84AI%E5%8A%9F%E8%83%BD%E9%9D%A9%E5%91%BD%EF%BC%8C%E5%8F%AF%E5%B0%86%E7%94%A8%E6%88%B7%E4%BD%BF%E7%94%A8AI%E7%9A%84%E4%B9%A0%E6%83%AF%E6%96%87%E4%BB%B6%E5%8C%96%E7%AE%A1%E7%90%86%E3%80%82%E5%AE%83%E8%83%BD%E8%A7%A3%E5%86%B3Claude%E5%81%A5%E5%BF%98%E3%80%81%E9%9C%80%E9%87%8D%E5%A4%8D%E6%8F%90%E7%A4%BA%E8%AF%8D%E7%9A%84%E9%97%AE%E9%A2%98%EF%BC%8C%E5%B0%86%E4%BB%BB%E5%8A%A1%E8%AF%B4%E6%98%8E%E3%80%81%E5%B7%A5%E5%85%B7%E4%BB%A3%E7%A0%81%E7%AD%89%E6%89%93%E5%8C%85%E6%88%90%20)[\[6\]](https://www.facebook.com/iamvista/photos/%E6%9F%90%E5%A4%A9%E6%B7%B1%E5%A4%9C%E6%88%91%E6%AD%A3%E5%9C%A8%E8%B6%95%E4%B8%80%E4%BB%BD%E6%96%87%E4%BB%B6%E5%A4%A9%E5%95%8A%E5%90%8C%E6%A8%A3%E7%9A%84%E6%9E%B6%E6%A7%8B%E5%90%8C%E6%A8%A3%E7%9A%84%E8%AA%9E%E6%B0%A3%E5%90%8C%E6%A8%A3%E7%9A%84%E6%A0%BC%E5%BC%8F%E8%A6%81%E6%B1%82%E4%BD%86%E6%88%91%E5%8F%88%E5%BE%97%E9%87%8D%E6%96%B0%E6%89%93%E4%B8%80%E6%AC%A1%E8%AB%8B%E7%94%A8%E9%80%99%E5%80%8B%E6%A0%BC%E5%BC%8F%E8%AB%8B%E5%85%88%E5%95%8F%E4%B8%89%E5%80%8B%E6%BE%84%E6%B8%85%E5%95%8F%E9%A1%8C%E8%AB%8B%E6%8A%8A%E8%BC%B8%E5%87%BA%E5%88%86%E6%88%90%E5%9B%9B%E6%AE%B5%E8%AB%8B%E9%99%84%E4%B8%8A%E5%8F%AF%E7%9B%B4%E6%8E%A5%E8%B2%BC%E5%88%B0-notion-%E7%9A%84/10162293093624053/#:~:text=,%E6%88%91%E6%80%8E%E9%BA%BC%E5%81%9A%E4%B8%80%E5%80%8B%E5%A5%BDskill%EF%BC%8C%E8%AE%8A%E6%88%90skill)。 **技能文件结构:** 每个Skill通常以独立文件夹形式存在,并包含一个核心定义文件(例如SKILL.md)以及可选的示例、模板和脚本子目录[\[7\]](https://ide.unitmesh.cc/spec/claude-skill#:~:text=%E6%8A%80%E8%83%BD%E7%BB%93%E6%9E%84%EF%BC%9A)。其基本结构如下(以my-skill为例)[\[8\]](https://ide.unitmesh.cc/spec/claude-skill#:~:text=my,%E8%BE%85%E5%8A%A9%E8%84%9A%E6%9C%AC%EF%BC%88%E5%8F%AF%E9%80%89%EF%BC%89): my-skill/ ├── SKILL.md # 技能定义(必需,含YAML元数据和Markdown内容) ├── examples/ # 示例文件(可选) ├── templates/ # 模板文件(可选) └── scripts/ # 辅助脚本(可选) 在SKILL.md中,开头使用YAML格式的**前置元数据(Skill Manifest)**描述技能信息,接着是Markdown格式的技能内容[\[9\]](https://ide.unitmesh.cc/spec/claude-skill#:~:text=%E6%8A%80%E8%83%BD%E5%AE%9A%E4%B9%89%E6%A0%BC%E5%BC%8F)。例如: \--- name: my-skill-name description: 用一句话清晰描述此技能的功能和适用场景 version: 1.0.0 author: 张三 tags: \[类别1, 类别2\] \--- \# 我的技能名称
\## 目的 这里解释该技能要解决的问题、使用场景。
\## 指令 这里编写赋予AI的详细指令,指导其执行本技能涉及的任务步骤。
\## 示例 \### 示例 1:基本用法 输入:... 输出:...
\### 示例 2:进阶用法 输入:... 输出:...
\## 指南 \- 提示使用者的一些注意事项或最佳实践要点。
\## 限制 \- 列出本技能的已知局限,如不会处理某些特殊情况等。 上述示例展示了技能定义模板的关键部分[\[10\]](https://ide.unitmesh.cc/spec/claude-skill#:~:text=,%E7%B1%BB%E5%88%AB2)[\[11\]](https://ide.unitmesh.cc/spec/claude-skill#:~:text=):YAML区块定义技能名称、描述、版本、作者和标签等元数据;随后Markdown正文按章节提供了技能的**用途说明**、AI执行的**详细指令**、使用**示例**以及指导和限制。元数据使技能模块**可版本化**和**可审计**,而指令部分则充当AI执行该技能的"剧本"。此外,开发者可以在scripts/目录中加入辅助脚本(如Python脚本)或在templates/加入模板文件,供AI在执行技能时调用[\[12\]](https://ide.unitmesh.cc/spec/claude-skill#:~:text=Anthropic%20Claude%20Skills%20%E6%98%AF%E5%8C%85%E5%90%AB%E6%8C%87%E4%BB%A4%E3%80%81%E8%84%9A%E6%9C%AC%E5%92%8C%E8%B5%84%E6%BA%90%E7%9A%84%E6%96%87%E4%BB%B6%E5%A4%B9%EF%BC%8CClaude%20%E5%8A%A8%E6%80%81%E5%8A%A0%E8%BD%BD%E4%BB%A5%E6%8F%90%E9%AB%98%E4%B8%93%E4%B8%9A%E4%BB%BB%E5%8A%A1%E7%9A%84%E6%80%A7%E8%83%BD%E3%80%82)[\[13\]](https://ide.unitmesh.cc/spec/claude-skill#:~:text=)。这种结构使得每个Skill成为**可组合、可共享**的最小单元,方便团队协作和知识复用[\[14\]](https://github.com/wensia/xiaohongshu-poster-generator#:~:text=wensia%2Fxiaohongshu,Code%20%2B%20MCP%20%E7%9A%84%E5%B0%8F%E7%BA%A2%E4%B9%A6%E6%98%9F%E5%BA%A7%E5%86%85%E5%AE%B9%E8%87%AA%E5%8A%A8%E5%8C%96%E5%B7%A5%E4%BD%9C%E6%B5%81%EF%BC%8C%E6%94%AF%E6%8C%81%E6%B5%B7%E6%8A%A5%E7%94%9F%E6%88%90%E3%80%81%E7%88%86%E6%96%87%E5%88%86%E6%9E%90%E3%80%81%E5%86%85%E5%AE%B9%E5%8F%91%E5%B8%83%E7%AD%89%E5%8A%9F%E8%83%BD%E3%80%82%20%E5%8A%9F%E8%83%BD%E7%89%B9%E6%80%A7)。 **构建Skills的步骤(从入门到精通):** 对于初学者,推荐循序渐进掌握Claude Skills的制作与应用,可参考以下三步策略[\[15\]](https://blog.csdn.net/yangshangwei/article/details/156836796#:~:text=%E4%B8%80%E4%B8%AA%E5%8F%AF%E8%90%BD%E5%9C%B0%E7%9A%84%E4%B8%89%E6%AD%A5%E6%B3%95,3%20%E7%AC%AC%E4%B8%89%E6%AD%A5%EF%BC%9A%E6%8A%8AClaude%20%E5%BD%93%E2%80%9C%E6%90%AD%E5%AD%90%E2%80%9D%EF%BC%8C%E8%80%8C%E4%B8%8D%E6%98%AF%E6%90%9C%E7%B4%A2%E6%A1%86): - **直接复用现有技能:** **"拿来即用"**是上手Claude Skills最快的方法[\[15\]](https://blog.csdn.net/yangshangwei/article/details/156836796#:~:text=%E4%B8%80%E4%B8%AA%E5%8F%AF%E8%90%BD%E5%9C%B0%E7%9A%84%E4%B8%89%E6%AD%A5%E6%B3%95,3%20%E7%AC%AC%E4%B8%89%E6%AD%A5%EF%BC%9A%E6%8A%8AClaude%20%E5%BD%93%E2%80%9C%E6%90%AD%E5%AD%90%E2%80%9D%EF%BC%8C%E8%80%8C%E4%B8%8D%E6%98%AF%E6%90%9C%E7%B4%A2%E6%A1%86)。Anthropic官方提供了丰富的内置技能库(涵盖文档处理、代码开发、创意生成等),社区也涌现出大量共享技能包。初始阶段,先尝试调用这些**官方和社区现成Skills**来完成任务。例如使用/skill.docx快速生成Word报告,或/skill.xlsx分析Excel数据等[\[16\]](https://ide.unitmesh.cc/spec/claude-skill#:~:text=,%E9%A2%84%E7%AE%97%E6%98%8E%E7%BB%86%20%E5%90%AF%E7%94%A8%E4%BF%AE%E8%AE%A2%E8%B7%9F%E8%B8%AA%E4%BB%A5%E4%BE%9B%E5%AE%A1%E6%9F%A5)[\[17\]](https://ide.unitmesh.cc/spec/claude-skill#:~:text=)。通过观察这些技能的效果和实现方式,来体会技能的结构和用途。 - **将重复任务技能化:** 当发现某类任务需要频繁向AI重复说明时,就可以考虑将其封装为自定义Skill[\[15\]](https://blog.csdn.net/yangshangwei/article/details/156836796#:~:text=%E4%B8%80%E4%B8%AA%E5%8F%AF%E8%90%BD%E5%9C%B0%E7%9A%84%E4%B8%89%E6%AD%A5%E6%B3%95,3%20%E7%AC%AC%E4%B8%89%E6%AD%A5%EF%BC%9A%E6%8A%8AClaude%20%E5%BD%93%E2%80%9C%E6%90%AD%E5%AD%90%E2%80%9D%EF%BC%8C%E8%80%8C%E4%B8%8D%E6%98%AF%E6%90%9C%E7%B4%A2%E6%A1%86)。经验法则是:**凡是你在对话中重复3次以上的提示**,都值得沉淀为技能模块。首先,整理该任务的明确步骤和专业知识要点,把你的**方法论**写下来[\[6\]](https://www.facebook.com/iamvista/photos/%E6%9F%90%E5%A4%A9%E6%B7%B1%E5%A4%9C%E6%88%91%E6%AD%A3%E5%9C%A8%E8%B6%95%E4%B8%80%E4%BB%BD%E6%96%87%E4%BB%B6%E5%A4%A9%E5%95%8A%E5%90%8C%E6%A8%A3%E7%9A%84%E6%9E%B6%E6%A7%8B%E5%90%8C%E6%A8%A3%E7%9A%84%E8%AA%9E%E6%B0%A3%E5%90%8C%E6%A8%A3%E7%9A%84%E6%A0%BC%E5%BC%8F%E8%A6%81%E6%B1%82%E4%BD%86%E6%88%91%E5%8F%88%E5%BE%97%E9%87%8D%E6%96%B0%E6%89%93%E4%B8%80%E6%AC%A1%E8%AB%8B%E7%94%A8%E9%80%99%E5%80%8B%E6%A0%BC%E5%BC%8F%E8%AB%8B%E5%85%88%E5%95%8F%E4%B8%89%E5%80%8B%E6%BE%84%E6%B8%85%E5%95%8F%E9%A1%8C%E8%AB%8B%E6%8A%8A%E8%BC%B8%E5%87%BA%E5%88%86%E6%88%90%E5%9B%9B%E6%AE%B5%E8%AB%8B%E9%99%84%E4%B8%8A%E5%8F%AF%E7%9B%B4%E6%8E%A5%E8%B2%BC%E5%88%B0-notion-%E7%9A%84/10162293093624053/#:~:text=,%E6%88%91%E6%80%8E%E9%BA%BC%E5%81%9A%E4%B8%80%E5%80%8B%E5%A5%BDskill%EF%BC%8C%E8%AE%8A%E6%88%90skill)。然后,利用Claude Code内置的**Skill Creator**功能,让AI协助你生成技能骨架[\[18\]](https://hbwdj.gov.cn/appbdetail-imqqsmrp9897358.d#:~:text=%E9%AA%97%E4%BD%A0%E7%9A%84%EF%BC%8C%E5%85%B6%E5%AE%9EAI%E6%A0%B9%E6%9C%AC%E4%B8%8D%E9%9C%80%E8%A6%81%E9%82%A3%E4%B9%88%E5%A4%9A%E6%8F%90%E7%A4%BA%E8%AF%8D%E3%80%82%20%E4%BD%A0%E5%8F%AA%E9%9C%80%E8%A6%81%E8%B0%83%E7%94%A8AI%20%E6%9C%AC%E8%BA%AB%E7%9A%84%E2%80%9CSkill%20Creator%E2%80%9D%E6%8A%80%E8%83%BD%EF%BC%8C%E7%94%A8%E4%BD%A0%E7%9A%84%E8%AF%AD%E8%A8%80%E6%8F%8F%E8%BF%B0%E8%87%AA%E5%B7%B1%E7%9A%84%E9%9C%80%E6%B1%82%EF%BC%8C%E8%AE%A9AI%E8%87%AA%E5%8A%A8%E5%B8%AE%E4%BD%A0%E7%94%9F%E6%88%90%E4%B8%80%E9%97%A8%E6%8A%80%E8%83%BD%EF%BC%8C%E4%BD%BF%E7%94%A8%E8%B5%B7%E6%9D%A5%E9%9D%9E%E5%B8%B8%E5%8F%8B%E5%A5%BD%EF%BC%8CAI%E4%BC%9A%E4%B8%80%E6%AD%A5%E6%AD%A5%E5%BC%95%E5%AF%BC%E4%BD%A0%E8%AF%B4%E5%87%BA%E4%BD%A0%E7%9A%84%E9%9C%80%E6%B1%82%EF%BC%8C%E4%BD%A0%E5%8F%AA%20)。例如,你可以对Claude说:"帮我创建一个Skill,用于根据文章内容自动选择并插入配图"。Claude会逐步引导你提供细节,自动产出包含正确YAML元数据和指令内容的Skill文件[\[18\]](https://hbwdj.gov.cn/appbdetail-imqqsmrp9897358.d#:~:text=%E9%AA%97%E4%BD%A0%E7%9A%84%EF%BC%8C%E5%85%B6%E5%AE%9EAI%E6%A0%B9%E6%9C%AC%E4%B8%8D%E9%9C%80%E8%A6%81%E9%82%A3%E4%B9%88%E5%A4%9A%E6%8F%90%E7%A4%BA%E8%AF%8D%E3%80%82%20%E4%BD%A0%E5%8F%AA%E9%9C%80%E8%A6%81%E8%B0%83%E7%94%A8AI%20%E6%9C%AC%E8%BA%AB%E7%9A%84%E2%80%9CSkill%20Creator%E2%80%9D%E6%8A%80%E8%83%BD%EF%BC%8C%E7%94%A8%E4%BD%A0%E7%9A%84%E8%AF%AD%E8%A8%80%E6%8F%8F%E8%BF%B0%E8%87%AA%E5%B7%B1%E7%9A%84%E9%9C%80%E6%B1%82%EF%BC%8C%E8%AE%A9AI%E8%87%AA%E5%8A%A8%E5%B8%AE%E4%BD%A0%E7%94%9F%E6%88%90%E4%B8%80%E9%97%A8%E6%8A%80%E8%83%BD%EF%BC%8C%E4%BD%BF%E7%94%A8%E8%B5%B7%E6%9D%A5%E9%9D%9E%E5%B8%B8%E5%8F%8B%E5%A5%BD%EF%BC%8CAI%E4%BC%9A%E4%B8%80%E6%AD%A5%E6%AD%A5%E5%BC%95%E5%AF%BC%E4%BD%A0%E8%AF%B4%E5%87%BA%E4%BD%A0%E7%9A%84%E9%9C%80%E6%B1%82%EF%BC%8C%E4%BD%A0%E5%8F%AA%20)。接着你可以在此基础上完善调整,加入脚本或模板,实现更复杂的逻辑。通过这种**人机协作**,即使不熟悉YAML语法的新手也能快速构建出可用的技能雏形。 - **测试迭代与高级优化:** 创建Skill后,在实际项目中多次调用测试,验证其稳定性和效果。根据AI每次执行技能的表现,不断**迭代优化**指令和脚本:完善边缘情况处理,增加更多示例和指南,让AI对技能意图理解更准确。Anthropic提供了**渐进式构建**技能的最佳实践,例如"MASTER六步法"等框架,用于从手工Prompt打磨到自动技能的完整流程[\[19\]](https://lilys.ai/zh/notes/claude-code-20251031/claude-code-skills-automate-everything#:~:text=%E6%83%B3%20%E8%87%AA%E5%8A%A8%E5%8C%96%E4%BD%A0%E6%89%80%E5%81%9A%E7%9A%84%E4%B8%80%E5%88%87%20%E5%90%97%EF%BC%9F%E5%AD%A6%E4%B9%A0%E5%A6%82%E4%BD%95%E5%88%A9%E7%94%A8%20%E5%85%8B%E5%8A%B3%E5%BE%B7%C2%B7%E7%A7%91%E5%BE%B7%E6%8A%80%E8%83%BD%20%E5%88%9B%E5%BB%BA%E8%87%AA%E5%AE%9A%E4%B9%89%E5%B7%A5%E4%BD%9C%E6%B5%81%E3%80%82%E6%9C%AC%E5%86%85%E5%AE%B9%E6%8F%AD%E7%A4%BA%E4%BA%86,%E5%85%AD%E6%AD%A5MASTER%E6%A1%86%E6%9E%B6%EF%BC%8C%E6%95%99%E4%BD%A0%E5%A6%82%E4%BD%95%E9%80%9A%E8%BF%87%20%E8%BF%AD%E4%BB%A3%E5%8F%8D%E9%A6%88%20%E8%AE%AD%E7%BB%83AI%E3%80%82%E6%8E%8C%E6%8F%A1%E6%AD%A4%E6%96%B9%E6%B3%95%EF%BC%8C%E4%BD%A0%E5%B0%B1%E8%83%BD%E5%B0%86%E9%87%8D%E5%A4%8D%E4%BB%BB%E5%8A%A1%E8%BD%AC%E5%8C%96%E4%B8%BA%E9%AB%98%E6%95%88%E7%9A%84%20%E7%B3%BB%E7%BB%9F%E5%8C%96%E6%8A%80%E8%83%BD%EF%BC%8C%E5%AE%9E%E7%8E%B0%E5%B7%A5%E4%BD%9C%E6%95%88%E7%8E%87%E7%9A%84%E6%8C%87%E6%95%B0%E7%BA%A7%E6%8F%90%E5%8D%87%E3%80%82)[\[20\]](https://lilys.ai/zh/notes/claude-code-20251031/claude-code-skills-automate-everything#:~:text=4.%20MASTER%E6%A1%86%E6%9E%B6%EF%BC%9A%E7%AC%AC%E4%BA%8C%E9%98%B6%E6%AE%B5%E2%80%94%E2%80%94%E7%B3%BB%E7%BB%9F%E5%8C%96%E4%B8%BA%E6%8A%80%E8%83%BD%20)。核心思想是在对话中**演示和纠正**AI完成任务的全过程,然后一句话让Claude将过程固化为新技能[\[20\]](https://lilys.ai/zh/notes/claude-code-20251031/claude-code-skills-automate-everything#:~:text=4.%20MASTER%E6%A1%86%E6%9E%B6%EF%BC%9A%E7%AC%AC%E4%BA%8C%E9%98%B6%E6%AE%B5%E2%80%94%E2%80%94%E7%B3%BB%E7%BB%9F%E5%8C%96%E4%B8%BA%E6%8A%80%E8%83%BD%20)[\[21\]](https://lilys.ai/zh/notes/claude-code-20251031/claude-code-skills-automate-everything#:~:text=1.%20%E7%B3%BB%E7%BB%9F%E5%8C%96%E7%9B%AE%E6%A0%87%EF%BC%9A%E5%B0%86%E5%AD%A6%E5%88%B0%E7%9A%84%E6%89%80%E6%9C%89%E5%85%B3%E9%94%AE%E7%BB%8F%E9%AA%8C%E6%95%99%E8%AE%AD%E8%BD%AC%E5%8C%96%E4%B8%BA%20%E7%B3%BB%E7%BB%9F%E5%8C%96%E6%8A%80%E8%83%BD%E3%80%82,48)。之后再让Claude应用该技能执行类似任务,观察输出并调整技能定义[\[22\]](https://lilys.ai/zh/notes/claude-code-20251031/claude-code-skills-automate-everything#:~:text=match%20at%20L3327%205,57)。这种循环可以逐步提升技能质量,直到达到"专家级"水平。在进阶阶段,你还可以尝试**组合多个Skills**协同工作,以及利用Claude Code提供的Hooks和子代理机制扩展技能的功能边界,实现更复杂的自动化开发流程。 通过以上步骤,从模仿官方示例到定制自己的技能,再到深入技能架构的优化,你将完成从入门到精通Claude Code Skills的蜕变。在这一过程中,要牢记技能开发的目的:**减少重复劳动、固化专业知识,并让AI行为更可控**。下一节我们将进一步探讨这一新范式如何将开发流程从传统Prompt工程升级为模块化的自动化开发。 ## 从Prompt工程到可复用模块:范式转移 Claude Skills的出现标志着AI应用开发从"Prompt Engineering"(提示词工程)向"**Context Engineering**"(上下文工程)的范式转移[\[23\]](https://claudecn.com/blog/claude-skills-architecture/#:~:text=1)。传统的Prompt工程往往依赖人工反复调试长提示语,把所有指导都硬编码在一个巨大prompt里,不仅上下文窗口占用大,而且难以复用和维护。相反,Skills将复杂提示拆解为**可版本化、可审计、可组合**的**运行时模块**[\[24\]](https://claudecn.com/blog/claude-skills-architecture/#:~:text=%E4%B8%BA%E4%BA%86%E9%81%BF%E5%85%8D%E2%80%9C%E4%BB%8B%E7%BB%8D%E5%8A%9F%E8%83%BD%E4%BD%86%E7%BC%BA%E5%B0%91%E5%8F%AF%E6%A3%80%E9%AA%8C%E7%BB%93%E8%AE%BA%E2%80%9D%EF%BC%8C%E6%9C%AC%E6%96%87%E5%85%88%E6%8A%8A%20Skills%20%E7%9A%84%E6%9E%B6%E6%9E%84%E4%BB%B7%E5%80%BC%E6%94%B6%E6%95%9B%E6%88%90%E4%B8%80%E5%8F%A5%E5%8F%AF%E9%AA%8C%E8%AF%81%E7%9A%84%E5%91%BD%E9%A2%98%EF%BC%9A)。这一转变带来了三大核心收益[\[25\]](https://claudecn.com/blog/claude-skills-architecture/#:~:text=Skills%20%E6%8A%8A%20LLM%20%E7%B3%BB%E7%BB%9F%E4%BB%8E%E2%80%9C%E6%96%87%E6%9C%AC%E5%A0%86%E5%8F%A0%E7%9A%84%E5%8D%95%E4%BD%93%E6%8F%90%E7%A4%BA%E8%AF%8D%E2%80%9D%EF%BC%8C%E9%87%8D%E6%9E%84%E4%B8%BA%E2%80%9C%E5%8F%AF%E7%89%88%E6%9C%AC%E5%8C%96%E3%80%81%E5%8F%AF%E5%AE%A1%E8%AE%A1%E3%80%81%E5%8F%AF%E7%BB%84%E5%90%88%E7%9A%84%E8%BF%90%E8%A1%8C%E6%97%B6%E6%A8%A1%E5%9D%97%E2%80%9D%EF%BC%9B%E6%A0%B8%E5%BF%83%E6%94%B6%E7%9B%8A%E6%9D%A5%E8%87%AA%E4%B8%89%E4%BB%B6%E4%BA%8B%EF%BC%9A%E4%B8%8A%E4%B8%8B%E6%96%87%E9%A2%84%E7%AE%97%E5%8F%AF%E6%8E%A7%E3%80%81%E6%89%A7%E8%A1%8C%E8%B7%AF%E5%BE%84%E5%8F%AF%E6%8E%A7%E3%80%81%E6%9D%83%E9%99%90%E8%BE%B9%E7%95%8C%E5%8F%AF%E6%8E%A7%E3%80%82): - **上下文预算可控:** Skills采用**分层渐进披露**机制,将提示内容分为元数据、指令、资源三层,按需加载[\[26\]](https://claudecn.com/blog/claude-skills-architecture/#:~:text=,3.%20%E6%8A%80%E6%9C%AF%E8%A7%A3%E6%9E%84%EF%BC%9ASkill%20%E7%9A%84%E7%89%A9%E7%90%86%E5%BD%A2%E6%80%81%E4%B8%8E%E8%A7%84%E8%8C%83)。模型初始只载入轻量的元信息(如技能名称和描述),在确认相关任务时再注入具体指令,必要时最后才加载代码模板或数据资源[\[3\]](https://claudecn.com/#:~:text=Agent%20Skills%20%E5%8F%AF%E5%A4%8D%E7%94%A8%E7%9A%84%E6%8A%80%E8%83%BD%E7%B3%BB%E7%BB%9F%EF%BC%8C%E8%AE%A9%20Claude%20%E6%8E%8C%E6%8F%A1%E7%89%B9%E5%AE%9A%E9%A2%86%E5%9F%9F%E4%B8%93%E4%B8%9A%E7%9F%A5%E8%AF%86%E3%80%82%E5%8C%85%E6%8B%AC,%E8%BF%9E%E6%8E%A5%E5%A4%96%E9%83%A8%E5%B7%A5%E5%85%B7%E5%92%8C%E6%95%B0%E6%8D%AE%E6%BA%90%EF%BC%8C%E6%89%A9%E5%B1%95%20Claude%20%E7%9A%84%E8%83%BD%E5%8A%9B%E8%BE%B9%E7%95%8C%EF%BC%8C%E6%9E%84%E5%BB%BA%E5%BC%BA%E5%A4%A7%E7%9A%84%20AI%20%E5%BA%94%E7%94%A8%E7%94%9F%E6%80%81%E3%80%82)。这样避免了"一次性把所有提示堆入上下文"的低效做法,防止上下文窗口的"公地悲剧"[\[27\]](https://claudecn.com/blog/claude-skills-architecture/#:~:text=,2.3%20%E6%8A%8A%E2%80%9C%E5%88%86%E5%B1%82%E6%8A%AB%E9%9C%B2%E2%80%9D%E6%8F%90%E5%8D%87%E4%B8%BA%E8%BF%90%E8%A1%8C%E6%97%B6%E7%8A%B6%E6%80%81%E6%9C%BA)。通过精细控制常驻、激活、执行三类上下文开销,开发者可以**显著降低Token占用**,将有限的上下文预算用在刀刃上。 - **执行路径可控:** 在Skills架构下,模型不再承担所有推理细节,而更像一个**编排者**。我们可以把复杂决策逻辑迁移到Skill的脚本中,由可测试的代码或明确的规则去执行,Claude则根据指令**调度脚本和资源**[\[28\]](https://claudecn.com/blog/claude-skills-architecture/#:~:text=,%E6%9D%83%E9%99%90%E8%BE%B9%E7%95%8C%E5%8F%AF%E6%8E%A7%EF%BC%9A%E7%94%A8%E6%B2%99%E7%AE%B1%E3%80%81%E7%BD%91%E7%BB%9C%E4%BB%A3%E7%90%86%E4%B8%8E%E6%9D%83%E9%99%90%E6%8F%90%E7%A4%BA%EF%BC%8C%E6%8A%8A%E5%B7%A5%E5%85%B7%E6%89%A7%E8%A1%8C%E9%9D%A2%E6%94%B6%E6%95%9B%E5%88%B0%E5%8F%AF%E5%AE%A1%E8%AE%A1%E3%80%81%E5%8F%AF%E6%B2%BB%E7%90%86%E7%9A%84%E8%BE%B9%E7%95%8C%E5%86%85%E3%80%82)。通过这种软硬结合,AI生成过程更具确定性,可重复性更高。例如一个技能可以包含预定义的正则脚本来处理文本格式,模型只需调用而非每次重新"即兴发挥"。这让AI行为如同运行程序模块一样**可预测、可验证**。 - **权限边界可控:** Skills配合Claude Code提供的沙箱和权限机制,可将AI的操作限制在安全边界内[\[28\]](https://claudecn.com/blog/claude-skills-architecture/#:~:text=,%E6%9D%83%E9%99%90%E8%BE%B9%E7%95%8C%E5%8F%AF%E6%8E%A7%EF%BC%9A%E7%94%A8%E6%B2%99%E7%AE%B1%E3%80%81%E7%BD%91%E7%BB%9C%E4%BB%A3%E7%90%86%E4%B8%8E%E6%9D%83%E9%99%90%E6%8F%90%E7%A4%BA%EF%BC%8C%E6%8A%8A%E5%B7%A5%E5%85%B7%E6%89%A7%E8%A1%8C%E9%9D%A2%E6%94%B6%E6%95%9B%E5%88%B0%E5%8F%AF%E5%AE%A1%E8%AE%A1%E3%80%81%E5%8F%AF%E6%B2%BB%E7%90%86%E7%9A%84%E8%BE%B9%E7%95%8C%E5%86%85%E3%80%82)。例如,在技能元数据或Claude配置中声明此技能允许的文件读写范围、外部命令白名单等,Claude执行技能时就不会越权[\[28\]](https://claudecn.com/blog/claude-skills-architecture/#:~:text=,%E6%9D%83%E9%99%90%E8%BE%B9%E7%95%8C%E5%8F%AF%E6%8E%A7%EF%BC%9A%E7%94%A8%E6%B2%99%E7%AE%B1%E3%80%81%E7%BD%91%E7%BB%9C%E4%BB%A3%E7%90%86%E4%B8%8E%E6%9D%83%E9%99%90%E6%8F%90%E7%A4%BA%EF%BC%8C%E6%8A%8A%E5%B7%A5%E5%85%B7%E6%89%A7%E8%A1%8C%E9%9D%A2%E6%94%B6%E6%95%9B%E5%88%B0%E5%8F%AF%E5%AE%A1%E8%AE%A1%E3%80%81%E5%8F%AF%E6%B2%BB%E7%90%86%E7%9A%84%E8%BE%B9%E7%95%8C%E5%86%85%E3%80%82)。通过引入**工具使用权限提示**以及隔离执行环境(如MCP服务器、虚拟机沙箱),开发者可以防范AI执行系统命令或访问敏感数据时可能带来的风险[\[29\]](https://claudecn.com/blog/claude-skills-architecture/#:~:text=%2A%204.1%20%E7%9B%91%E7%9D%A3%E8%80%85,6.2%20%E5%8D%8F%E5%90%8C%E6%9E%B6%E6%9E%84%EF%BC%9ASkill%20%E4%BD%9C%E4%B8%BA%E7%BC%96%E6%8E%92%E8%80%85)。相比任由AI自由解释模糊指令,这种方式下每个技能的**作用域和副作用都是受控**的,大幅提高了AI应用的安全性和可靠性。 综上所述,模块化的Skills架构将LLM从以前的"提示词拼凑的巨石"解耦为若干**可管理的小单元**[\[24\]](https://claudecn.com/blog/claude-skills-architecture/#:~:text=%E4%B8%BA%E4%BA%86%E9%81%BF%E5%85%8D%E2%80%9C%E4%BB%8B%E7%BB%8D%E5%8A%9F%E8%83%BD%E4%BD%86%E7%BC%BA%E5%B0%91%E5%8F%AF%E6%A3%80%E9%AA%8C%E7%BB%93%E8%AE%BA%E2%80%9D%EF%BC%8C%E6%9C%AC%E6%96%87%E5%85%88%E6%8A%8A%20Skills%20%E7%9A%84%E6%9E%B6%E6%9E%84%E4%BB%B7%E5%80%BC%E6%94%B6%E6%95%9B%E6%88%90%E4%B8%80%E5%8F%A5%E5%8F%AF%E9%AA%8C%E8%AF%81%E7%9A%84%E5%91%BD%E9%A2%98%EF%BC%9A)。开发者能够像搭建乐高一样组装技能,复用成熟模块,版本迭代升级技能库,并针对不同任务**按需加载**合适的技能组合[\[30\]](https://www.bilibili.com/opus/1132124852115734530#:~:text=Claude%20Skills%20%E5%AE%9E%E6%88%98%E6%8C%87%E5%8D%97%EF%BC%9A3%20%E5%88%86%E9%92%9F%E6%90%9E%E5%AE%9APPT%E3%80%81%E6%B5%B7%E6%8A%A5%E4%B8%8ELogo%20,%E5%AE%83%E5%B0%86%E5%B8%B8%E8%A7%81%E7%9A%84%E5%8A%9E%E5%85%AC%E4%BB%BB%E5%8A%A1%28%E5%A6%82Excel%20%E6%95%B0%E6%8D%AE%E5%88%86%E6%9E%90%E3%80%81PPT%20%E6%BC%94%E7%A4%BA%E7%94%9F%E6%88%90%E3%80%81%E6%96%87%E6%A1%A3%E5%A4%84%E7%90%86%E3%80%81%E5%93%81%E7%89%8C%E8%AE%BE%E8%AE%A1%E7%AD%89%29%E5%B0%81%E8%A3%85%E6%88%90%E5%8F%AF%E5%A4%8D%E7%94%A8%E7%9A%84%E6%8A%80%E8%83%BD%E6%A8%A1%E5%9D%97%E3%80%82%E8%BF%99%E7%A7%8D%E8%AE%BE%E8%AE%A1%E7%90%86%E5%BF%B5%E7%9A%84%E6%A0%B8%E5%BF%83%E4%BC%98%E5%8A%BF)[\[31\]](https://lilys.ai/zh/notes/claude-skills-20251022/no-code-ai-workflow-claude-skills#:~:text=Skill%20%E5%9F%BA%E7%A1%80%E7%BB%93%E6%9E%84%EF%BC%9A%E6%9C%80%E7%AE%80%E5%8D%95%E6%83%85%E5%86%B5%E4%B8%8B%EF%BC%8CSkill%20%E6%98%AF%E4%B8%80%E4%B8%AA%E5%8C%85%E5%90%ABSkill%20Markdown%20%E6%96%87%E4%BB%B6%E7%9A%84%E7%9B%AE%E5%BD%95,36%5D)。这不仅提高了开发效率,也让AI行为更加透明可控--我们可以审查每个Skill的定义,像审计代码一样检查AI决策依据。这种以**技能库**为中心的上下文工程理念,正在帮助AI从通用助手进化为各行业的专家[\[32\]](https://github.com/0xfnzero/AI-Code-Tutorials#:~:text=%E4%BB%8E%E9%9B%B6%E5%9F%BA%E7%A1%80%E5%88%B0%E9%AB%98%E7%BA%A7%E5%BA%94%E7%94%A8%EF%BC%8C%E7%B3%BB%E7%BB%9F%E5%AD%A6%E4%B9%A0Claude%20Code%EF%BC%8C%E6%8E%8C%E6%8F%A1AI%20%E8%BE%85%E5%8A%A9%E7%BC%96%E7%A8%8B%E6%8A%80%E8%83%BD%EF%BC%8C%E6%8F%90%E5%8D%87%E5%BC%80%E5%8F%91%E6%95%88%E7%8E%8710%20%E5%80%8D%20,md%E3%80%81%E5%B7%A5%E5%85%B7%E6%9D%83%E9%99%90%E3%80%81gh%20CLI%EF%BC%89%3B%20%E7%BB%99Claude%20%E6%9B%B4%E5%A4%9A%E5%B7%A5%E5%85%B7%EF%BC%88bash%E3%80%81MCP)。许多团队已开始构建自己的私有Skills库,将领域知识固化为技能手册,与团队共享使用[\[31\]](https://lilys.ai/zh/notes/claude-skills-20251022/no-code-ai-workflow-claude-skills#:~:text=Skill%20%E5%9F%BA%E7%A1%80%E7%BB%93%E6%9E%84%EF%BC%9A%E6%9C%80%E7%AE%80%E5%8D%95%E6%83%85%E5%86%B5%E4%B8%8B%EF%BC%8CSkill%20%E6%98%AF%E4%B8%80%E4%B8%AA%E5%8C%85%E5%90%ABSkill%20Markdown%20%E6%96%87%E4%BB%B6%E7%9A%84%E7%9B%AE%E5%BD%95,36%5D)。Prompt工程不再是黑箱中的玄学调参,而变成了**软件工程化**的过程:定义规范→编码技能→测试迭代→部署上线。可以预见,随着技能生态的完善和技能市场的兴起[\[33\]](https://news.qq.com/rain/a/20260107A02N2N00#:~:text=Skills%20%E5%B0%B1%E6%98%AF%E7%BB%99Claude%20%E7%9A%84),未来"会写Prompt"将不再是关键竞争力,取而代之的是"**会设计AI技能模块**",让AI真正成为持续进化的数字员工。 ## Claude/GPT-4 助力 Golang + Vue3 全栈项目开发 大型语言模型(LLM)如Claude 2和GPT-4已成为开发者强有力的**编程助手**。下面我们以Golang后端 + Vue3 (TypeScript + Vuetify UI) 前端的全栈项目为例,探讨如何利用Claude/GPT-4提高各环节开发效率,并重点说明**系统构建方法、提示词模板设计**及**工作流程集成**。 ### 系统构建与架构规划 在项目初期,AI可以充当"**架构顾问**"的角色,帮助选择技术栈、搭建骨架。开发者可以用自然语言向Claude/GPT-4描述项目需求和约束,让其建议合适的框架和方案。例如:"我想用Go开发后端REST API,用哪种Web框架比较合适?"Claude基于知识建议使用Gin框架,并解释其上手简单、配置少的优点[\[34\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=match%20at%20L93%20,side%20rendering%20and)。同样,对于前端,它可能推荐Vue3 + TypeScript配合Vuetify组件库,以快速构建响应式UI[\[35\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=match%20at%20L96%20,No%20magic%2C%20no)[\[36\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=,No%20magic%2C%20no)。在实际案例中,开发者Korbinian Schleifer分享道:他询问Claude选择Go框架,Claude推荐了Gin,使他很快建立起第一个路由;选择前端则采用Vue3 + TS,Claude也支持这一选择,认为Vue的模板语法更贴近HTML,学习曲线相对平缓[\[34\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=match%20at%20L93%20,side%20rendering%20and)[\[37\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=,No%20magic%2C%20no)。借助AI的建议,可以较快确定技术栈并初始化项目结构(例如生成基础的Go项目模块和Vue前端脚手架)。 **分层实现:** 拆解全栈系统时,建议先让AI协助规划前后端接口契约和模块划分。例如,可以让Claude梳理"后台需要提供哪些REST API,以及前端各页面对应哪些组件和状态管理"。通过逐步追问,Claude能够产出一个简单的**架构清单或示意图**。有趣的是,Claude擅长**文字描述架构**,甚至可以生成架构图的代码(如Excalidraw的JSON)供开发者直接渲染出图表[\[38\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=Claude%20can%20describe%20architecture%20well%2C,com%20which%20I%20then)。这有助于开发者和AI达成对系统设计的一致理解。 ### 提示词设计与编程对话技巧 在与AI pair programming(结对编程)的过程中,**提示词的设计**直接影响AI产出的质量和可控性。以下是实战中总结的Prompt技巧: - **提供充分的上下文:** _"Context is king."_ 与LLM协作编程时,不要吝啬提供背景信息[\[39\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=match%20at%20L316%20Give%20Claude,context%2C%20the%20better%20the%20suggestions)。明确告知AI当前项目的**技术栈、代码结构、已有代码片段**、使用的库版本,以及你已经尝试过的方法等[\[39\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=match%20at%20L316%20Give%20Claude,context%2C%20the%20better%20the%20suggestions)。上下文越完整,AI越能理解需求,给出针对性的建议。例如,在请求AI编写某个Vue组件时,先说明项目使用Vuetify版本、已有的全局样式等,AI就能生成更符合项目风格的代码。 - **过程式分解任务:** 避免一次让AI生成庞大复杂的代码,不妨将需求拆成**多步对话**。你可以先让AI**列出实现思路**或步骤清单,然后确认方案后再逐步让其实现每一步。这类似于指导一个新手程序员:先讨论方案,再Coding。Claude等模型擅长这种渐进式对话,会根据前文步骤逐一输出相应代码。在这个过程中,可灵活插入自己的想法,例如:"步骤2很棒,但请在实现时使用Vuetify的Grid系统。" 通过逐步细化指令,模型生成的代码将更符合预期。 - **及时反馈和明确决策:** 当AI给出多个方案或不确定措辞时,务必**明确告诉它你选择了哪一种**[\[40\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=match%20at%20L322%20When%20Claude,things%20you%E2%80%99ve%20already%20decided%20against)。例如Claude可能提出两种API设计思路,选定其中一个后应回复:"我决定采用方案B,接下来按照这个方案继续。" 这样AI在后续对话中会聚焦于你选定的方向,避免反复纠结已否定的路线[\[40\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=match%20at%20L322%20When%20Claude,things%20you%E2%80%99ve%20already%20decided%20against)。同样,如果AI生成的代码不符合预期,要迅速指出问题所在(bug、风格偏差或逻辑错误)并要求修改。通过这种**紧反馈回路**,AI相当于接受了快速code review,能更快调整输出方向。 - **保持风格一致性:** 对于多人协作或涉及特定编码规范的项目,可以在提示中强调代码风格和约定。例如:"请使用Go语言的标准错误处理模式,前端TS代码请遵循项目已有的ESLint规则,组件命名采用PascalCase。"Claude/GPT-4会据此自适应输出格式。实践表明,让AI扮演特定角色有助于风格统一,比如提示它"你是资深Go工程师"或"充当代码审核者",使其回答更严格遵循专业规范[\[41\]](https://www.cnblogs.com/treasury-manager/p/19217990#:~:text=OpenCode%20%E4%B8%80%E4%B8%AA%E7%A5%9E%E5%A5%87%E7%9A%84CLI%20%EF%BC%88%E5%8F%AF%E4%BB%A5%E8%9E%8D%E5%90%88Claude%20Code%2C%20iFLow,code%29%20%E2%86%90%20%E8%87%AA%E5%8A%A8%E4%BD%BF%E7%94%A8oracle%20%E7%9A%84%E6%A8%A1%E5%9E%8B%E2%86%93%20%E8%B0%83%E7%94%A8%40explore)。此外,一些高级用户会维护一个**项目说明文档**(如CLAUDE.md或README),里面列出项目编码准则,然后在每次对话开始时提供给AI参考。这相当于建立上下文的**长期记忆**。 - **控制对话长度:** 在长时间对话或大型项目中,注意LLM的上下文窗口限制。如果对话持续很多轮且内容庞杂,Claude可能出现响应变慢甚至遗忘前文细节的情况[\[42\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=After%20long%20chats%2C%20Claude%20noticeably,you%20have%20done%20so%20far)。此时有两个对策:一是**总结当前进度**,开启一个新会话,将概要和关键代码片段提供给新对话作为背景,让AI轻装上阵接着干[\[42\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=After%20long%20chats%2C%20Claude%20noticeably,you%20have%20done%20so%20far);二是利用像OpenCode这样的工具,它可以自动维护项目知识(通过索引代码库等),减轻每次都传输大量历史的负担[\[43\]](https://opencode.ai/docs#:~:text=%2Finit)[\[44\]](https://opencode.ai/docs#:~:text=You%20can%20ask%20OpenCode%20to,explain%20the%20codebase%20to%20you)。总之,要避免一口气把所有内容都塞给模型,多用**小步快跑、阶段重启**的方式确保AI始终聚焦有效信息。 通过上述Prompt工程技巧,开发者在Golang+Vue项目中与AI协作时,可以达到接近"**实时对话编程**"的体验。例如,有开发者分享他在几天内用Claude辅助完成了一个Go后端+Vue前端的CRUD应用,从中**学习了Go的新特性**(如 if err := ... 用法,Claude解释了其作用[\[45\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=if%20err%20%3A%3D%20godotenv,))也让Claude帮助生成了一些前端组件代码[\[46\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=match%20at%20L236%20This%20is,JavaScript%20pretending%20to%20be%20HTML)。期间他总结的心得是:"让Claude就像团队新人一样工作:给足资料,明确任务,迅速反馈,必要时重开小灶(新会话)。"[\[47\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=3,getting%20lost%20in%20tutorial%20hell)。结果显示,Claude在提供代码解释、框架用法指导和产生样板代码等方面极大加速了开发,但在某些细节上仍需要人工审查和调试。这提醒我们,AI虽然强大但不是万能,"Human in the loop"(人类介入)依然重要--**AI负责快,人工负责对**。利用好AI的长处(速度、记忆、生成能力)并辅以人类的判断力,才能在全栈开发中如虎添翼。 ### 工作流程集成与工具链结合 要充分发挥Claude/GPT-4在编码中的作用,建议将其集成进日常开发流程和工具链中,实现"AI协同开发"的闭环。以下是几种可行的集成方式: - **终端集成:** 使用AI驱动的命令行助手,如Anthropic官方的Claude Code CLI或者开源的OpenCode工具(详见下文),在终端中与AI互动编程[\[48\]](https://zhuanlan.zhihu.com/p/1991170184573122515#:~:text=OpenCode%20%E6%98%AF%E4%B8%80%E4%B8%AAAI%20%E7%BC%96%E7%A8%8B%E5%8A%A9%E6%89%8B%EF%BC%8C%E8%B7%91%E5%9C%A8%E4%BD%A0%E7%9A%84%E7%BB%88%E7%AB%AF%E9%87%8C%EF%BC%88%E5%B0%B1%E6%98%AF%E9%82%A3%E4%B8%AA%E9%BB%91%E8%89%B2%E7%AA%97%E5%8F%A3%EF%BC%89%E3%80%82%20%E4%BD%A0%E8%B7%9F%E5%AE%83%E8%AF%B4%E8%AF%9D%EF%BC%8C%E5%AE%83%E5%B0%B1%E5%B8%AE%E4%BD%A0%E5%86%99%E4%BB%A3%E7%A0%81%E3%80%82%20,%E2%86%92%20%E5%AE%83%E6%94%B9)。这类工具允许AI直接读取项目文件、执行Git命令、运行测试等。例如,你在终端对AI说"帮我实现登录功能",它就能新建代码文件、写入实现并运行测试验证功能[\[48\]](https://zhuanlan.zhihu.com/p/1991170184573122515#:~:text=OpenCode%20%E6%98%AF%E4%B8%80%E4%B8%AAAI%20%E7%BC%96%E7%A8%8B%E5%8A%A9%E6%89%8B%EF%BC%8C%E8%B7%91%E5%9C%A8%E4%BD%A0%E7%9A%84%E7%BB%88%E7%AB%AF%E9%87%8C%EF%BC%88%E5%B0%B1%E6%98%AF%E9%82%A3%E4%B8%AA%E9%BB%91%E8%89%B2%E7%AA%97%E5%8F%A3%EF%BC%89%E3%80%82%20%E4%BD%A0%E8%B7%9F%E5%AE%83%E8%AF%B4%E8%AF%9D%EF%BC%8C%E5%AE%83%E5%B0%B1%E5%B8%AE%E4%BD%A0%E5%86%99%E4%BB%A3%E7%A0%81%E3%80%82%20,%E2%86%92%20%E5%AE%83%E6%94%B9)。这种方式将AI无缝嵌入开发者熟悉的环境,大大减少在浏览器对话与IDE编辑器之间来回切换的摩擦。 - **IDE插件:** 如果偏好图形界面,可以使用支持GPT的IDE插件或扩展(如VS Code的Claude插件、ChatGPT Copilot等)。这些插件通常提供代码补全、文档查询、错误诊断等功能。例如在Vue组件文件中,选中一段报错代码,让AI解释错误原因并给出修改建议。集成在IDE中的AI还能结合LSP(语言服务器协议)获取更精准的代码上下文[\[49\]](https://github.com/anomalyco/opencode#:~:text=close%20and%20pricing%20will%20drop,what%27s%20possible%20in%20the%20terminal),从而提升回答专业度。许多现代编辑器插件已经支持将当前文件、甚至整个项目知识作为提示上下文传递给LLM,提高交互质量。 - **CI/CD 流程:** 在持续集成阶段,可以借助AI做代码审查和自动修复。比如Push代码后,触发一个CI Job调用Claude来分析最近的变更是否存在代码风格违背或潜在bug,然后自动提出修改建议(甚至直接开Pull Request修复简单问题)。Anthropic的Claude Code支持类似的**代码审查子代理**,可配置ESLint、单元测试覆盖率检查等,在每次提交时由AI代理提示改进之处[\[50\]](https://cloud.tencent.com/developer/article/2574516#:~:text=Claude%20Code%E4%BB%A3%E7%A0%81%E8%A7%84%E8%8C%83%E5%AE%88%E6%8A%A4%E8%80%85%E5%AD%90%E4%BB%A3%E7%90%86%E5%AE%9E%E6%88%98%E6%8C%87%E5%8D%97%20,Code%E4%BB%A3%E7%A0%81%E8%A7%84%E8%8C%83%E5%AD%90%E4%BB%A3%E7%90%86%E6%98%AFAI%E9%A9%B1%E5%8A%A8%E7%9A%84%E4%BB%A3%E7%A0%81%E8%B4%A8%E9%87%8F%E7%AE%A1%E5%AE%B6%EF%BC%8C%E8%83%BD%E8%87%AA%E5%8A%A8%E7%BB%9F%E4%B8%80%E5%9B%A2%E9%98%9F%E4%BB%A3%E7%A0%81%E9%A3%8E%E6%A0%BC%E3%80%81%E6%89%A7%E8%A1%8C%E5%91%BD%E5%90%8D%E8%A7%84%E8%8C%83%E3%80%81%E6%A3%80%E6%9F%A5%E6%B5%8B%E8%AF%95%E8%A6%86%E7%9B%96%E7%8E%87%E3%80%82%E9%80%9A%E8%BF%87ESLint%E3%80%81Prettier%E7%AD%89%E5%B7%A5%E5%85%B7%E9%85%8D%E7%BD%AE%EF%BC%8C)。这种AI辅助的代码守护者可以减轻Reviewer负担,保证代码质量的一致性。 - **文档和测试生成:** 将AI融入开发的"后勤"工作中也是一种高效做法。例如利用GPT-4根据代码自动生成文档注释、根据函数签名生成单元测试样例等。许多开源工具(如OpenAI's Codex、或OpenCode中的特定命令)可以一键生成注释和测试代码。开发者可以让AI先生成,再自行检查润色,从而快速补齐文档和测试,提高代码可维护性。 总的来说,无论通过何种途径集成AI助手,都应当**明确AI担当的角色**:是需求的分析者、编码的执行者还是结果的评审者。合理分配这些角色,让AI参与从需求->设计->编码->测试的各个环节,并建立**反馈闭环**(例如AI写完代码立即运行测试验证,失败再由AI修复),就能够形成流水线式的智能开发流程。在实践中,随着使用AI的成熟度提高,你甚至会把一些日常繁琐任务全权交给AI处理,例如批量重构代价不高的代码、格式标准化、依赖升级兼容性修改等,而自己专注于核心业务逻辑和架构决策。这正是AI辅助开发所追求的目标:**让人类做更高价值的创造,让AI承担重复冗杂的劳动**。 ## 开源AI编程工具集成:OpenCode 与 Oh My OpenCode 为了将大模型更好地融入实际开发,一些开源项目提供了强大的工具和插件。如前所述,OpenCode是当前炙手可热的开源AI编程助手,而"Oh My OpenCode (OMO)"则是其上一款明星插件,能将单Agent升级为多Agent协作。下面我们详细介绍两者的特点、使用方法和最佳实践。 ### OpenCode:终端AI编程助手 ![](data:image/png;base64,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) _OpenCode终端界面示例:开发者请求修改主页按钮颜色,AI在代码库中搜索相关文件并提出修改方案。界面底部显示当前模式为"Build",使用的模型是Claude Opus 4.5(Claude Code的API型号),以及OpenCode Zen推荐的模型配置。开发者可通过快捷键与AI交互(如Tab切换Agent模式)。该界面展示了OpenCode如何理解指令并进行代码操作的过程。_[_\[51\]_](https://github.com/anomalyco/opencode#:~:text=OpenCode%20includes%20two%20built,key)[_\[52\]_](https://opencode.ai/docs#:~:text=1) OpenCode是一个**完全开源的AI编码Agent**,支持在终端、桌面应用或IDE插件中运行[\[53\]](https://opencode.ai/docs#:~:text=OpenCode%20is%20an%20open%20source,desktop%20app%2C%20or%20IDE%20extension)。它的设计初衷是提供类似Anthropic Claude Code的体验,但**不绑定特定模型**且更加可定制[\[54\]](https://github.com/anomalyco/opencode#:~:text=%2A%20100,what%27s%20possible%20in%20the%20terminal)。OpenCode可以配置使用Anthropic Claude、OpenAI GPT-4、Google PaLM乃至本地开源模型,共支持超过75种LLM型号[\[55\]](https://github.com/anomalyco/opencode#:~:text=%2A%20100,push%20the%20limits%20of%20what%27s)。这种**模型无关性**意味着开发者可根据任务需要和成本自行选择AI引擎,而不局限于Claude,实现弹性升级[\[54\]](https://github.com/anomalyco/opencode#:~:text=%2A%20100,what%27s%20possible%20in%20the%20terminal)。另外,OpenCode内置对LSP(语言服务器协议)的支持,能自动根据项目语言加载相应语言服务器,增强代码理解能力[\[49\]](https://github.com/anomalyco/opencode#:~:text=close%20and%20pricing%20will%20drop,what%27s%20possible%20in%20the%20terminal)。 **安装与初始化:** 安装OpenCode非常简单,可通过npm、Homebrew等包管理器一键安装[\[56\]](https://github.com/anomalyco/opencode#:~:text=Installation)[\[57\]](https://github.com/anomalyco/opencode#:~:text=npm%20i%20,Any%20OS)。安装后,启动前需要提供LLM的API密钥并进行简单配置[\[58\]](https://opencode.ai/docs#:~:text=Configure)[\[59\]](https://opencode.ai/docs#:~:text=If%20you%20are%20new%20to,verified%20by%20the%20OpenCode%20team)。进入项目目录运行opencode,首次使用可执行命令/init让OpenCode扫描项目,生成AGENTS.md索引文件[\[60\]](https://opencode.ai/docs#:~:text=Next%2C%20initialize%20OpenCode%20for%20the,by%20running%20the%20following%20command)。这个文件记录了项目的结构和代码模式,有点像项目的"知识图谱",建议加入版本管理[\[61\]](https://opencode.ai/docs#:~:text=This%20will%20get%20OpenCode%20to,file%20in%20the%20project%20root)。有了它,OpenCode在对话中就能更好地引用项目文件、理解代码上下文。 **工作模式与命令:** OpenCode提供了**TUI交互界面**,你可以像与人聊天一样对它下指令。例如:"请在settings.ts中添加与notes.ts类似的身份验证代码"[\[62\]](https://opencode.ai/docs#:~:text=Make%20changes)。OpenCode会理解你的自然语言,自动完成在指定文件中插入代码的操作。其强大之处在于内置了**两种Agent模式**,可通过<kbd>TAB</kbd>键切换:[\[51\]](https://github.com/anomalyco/opencode#:~:text=OpenCode%20includes%20two%20built,key) - **Plan模式(规划模式)** - _"顾问"_:该模式下AI具有只读分析权限,不直接修改代码[\[51\]](https://github.com/anomalyco/opencode#:~:text=OpenCode%20includes%20two%20built,key)。当你提出新需求,OpenCode建议先按<kbd>TAB</kbd>切换到Plan模式,它会给出实现计划[\[63\]](https://opencode.ai/docs#:~:text=1)。例如你说"添加软删除功能",Plan模式下AI可能回复:"1)在数据库加deleted字段,2)后端新增恢复接口,3)前端加回收站页面…"[\[64\]](https://opencode.ai/docs#:~:text=Now%20let%E2%80%99s%20describe%20what%20we,want%20it%20to%20do)。规划模式会**禁用自动执行**,确保AI的提议经过你审阅。它甚至会在执行外部命令前征求同意(例如运行bash或安装依赖,需要你确认)[\[65\]](https://github.com/anomalyco/opencode#:~:text=%2A%20build%20,unfamiliar%20codebases%20or%20planning%20changes)。**最佳实践**是先让AI在Plan模式列出步骤和方案,开发者可以补充或更正细节[\[66\]](https://opencode.ai/docs#:~:text=2)。如觉得方案不完善,可继续对话调整直到满意为止[\[67\]](https://opencode.ai/docs#:~:text=2)。 - **Build模式(构建模式)** - _"执行者"_:在你接受规划后,按<kbd>TAB</kbd>回到Build模式,命令AI执行更改[\[68\]](https://opencode.ai/docs#:~:text=3)。Build模式下AI有**完全访问权限**,可以编辑文件、创建新文件、运行测试或Git命令等[\[51\]](https://github.com/anomalyco/opencode#:~:text=OpenCode%20includes%20two%20built,key)。延续上例,当你说"计划看起来不错,请动手实现",OpenCode将依次修改代码并可能运行项目验证[\[69\]](https://opencode.ai/docs#:~:text=Once%20you%20feel%20comfortable%20with,hitting%20the%20Tab%20key%20again)。它会将所有操作结果(如diff或者终端输出)显示在对话中,方便你跟踪进度。如果修改不符合预期,还可以及时喊停。**小贴士:**对于非常简单的修改,也可直接在Build模式下让AI改,无需先plan[\[62\]](https://opencode.ai/docs#:~:text=Make%20changes)--但前提是你对改动很有把握并已提供足够细节,否则直接构建可能出现偏差。 OpenCode在交互上还有几个贴心功能:你可以在对话中通过@文件名快速引用项目文件,AI会自动加载其内容作为上下文[\[70\]](https://opencode.ai/docs#:~:text=Tip);如果AI修改了文件,你可以用/undo命令**撤销更改**,一步步回滚到之前状态[\[71\]](https://opencode.ai/docs#:~:text=Let%E2%80%99s%20say%20you%20ask%20OpenCode,to%20make%20some%20changes);完成某个任务后,用/share可以生成对话的分享链接,方便团队其他成员查看这次AI改动的经过[\[72\]](https://opencode.ai/docs#:~:text=Share)。这些特性能鼓励开发者大胆尝试AI修改,因为**所有操作都可追溯和撤销**,相当于给AI上了一道"安全网"。 **最佳实践提示:** 使用OpenCode时,养成一些习惯可以大大提升体验: - **将AI当作新人**:对OpenCode的指令要清晰具体,必要时分解成子任务。它官方文档建议开发者"像对团队新人讲解任务那样与AI对话"[\[73\]](https://opencode.ai/docs#:~:text=You%20want%20to%20give%20OpenCode,junior%20developer%20on%20your%20team)。这意味着描述需求时要交代背景、"Done的定义"等,使AI少走弯路。 - **多给示例和引用**:如果要AI写类似已有代码的功能,引用相关文件路径并指出相似之处[\[74\]](https://opencode.ai/docs#:~:text=We%20need%20to%20add%20authentication,look%20at%20how%20this%20is)。例如:"参考notes.ts里的做法,在settings.ts实现类似逻辑"[\[75\]](https://opencode.ai/docs#:~:text=without%20having%20to%20review%20the,plan%20first)。OpenCode会据此调取对应文件,提高修改正确率。 - **审阅Plan并反馈**:Plan模式输出方案后务必仔细检查,每条步骤是否符合预期。如有遗漏或不妥,直接在对话中反馈修改。**不要吝惜反馈**--AI需要你的指引来校准方向[\[40\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=match%20at%20L322%20When%20Claude,things%20you%E2%80%99ve%20already%20decided%20against)。充分的前期沟通会减少后期反复。 - **小步提交**:AI每完成一部分改动,可先在本地运行测试或预览效果,再决定让其继续下一步。分阶段验收能及时发现问题,便于快速/undo回滚并调整Prompt。 OpenCode的出现,使个人开发者也能拥有"AI对话编程"的高效体验。有博文称其是"终端里的AI老司机",让写代码"爽到飞起"[\[76\]](https://blog.csdn.net/u012094427/article/details/148866474#:~:text=%E4%BB%8A%E5%A4%A9%E6%88%91%E4%BB%AC%E8%A6%81%E8%81%8A%E7%9A%84%E7%A1%AC%E6%A0%B8%E8%AF%9D%E9%A2%98%EF%BC%8C%E6%98%AF%E4%B8%AA%E8%AE%A9%E6%9E%81%E5%AE%A2%E4%BB%AC%E9%A2%A4%E6%8A%96%E3%80%81%E8%AE%A9%E7%A8%8B%E5%BA%8F%E5%91%98%E4%BB%AC%E5%B0%96%E5%8F%AB%EF%BC%8C%E8%AE%A9%E5%86%99%E4%BB%A3%E7%A0%81%E7%88%BD%E5%88%B0%E9%A3%9E%E8%B5%B7%E7%9A%84%E5%AD%98%E5%9C%A8%E2%80%94%E2%80%94OpenCode%EF%BC%8C%E5%BC%80%E6%BA%90AI%E7%BB%88%E7%AB%AF%E7%BC%96%E7%A0%81%E5%8A%A9%E6%89%8B%E3%80%82%20%E5%9C%A8AI%E5%85%A8%E8%83%BD%E5%86%99%E4%BB%A3%E7%A0%81%E3%80%81Copilot%E5%AE%B6%E5%A4%A7%E4%B8%9A%E5%A4%A7%E3%80%81%20)。而当OpenCode与更强大的插件Oh My OpenCode结合后,威力更是倍增--我们接下来就介绍OMO插件如何将AI辅助开发推向新的高度。 ### Oh My OpenCode:Sisyphus多代理增强插件 Oh My OpenCode(简称OMO)是OpenCode的明星社区插件,由开发者Yeongyu Kim开源。它被称为"**最强Agent挂架**(Agent Harness)",内部代号"Sisyphus",寓意让AI像西西弗斯般不懈耕耘,为你完成复杂开发任务[\[77\]](https://x.com/Nateemerson/status/2002043382953288046/photo/1#:~:text=YeonGyu,and%20practices%20for%20agentic%20coding)[\[78\]](https://github.com/code-yeongyu/oh-my-opencode#:~:text=The%20Best%20Agent%20Harness,Agent%20that%20codes%20like%20you)。简单来说,OMO通过引入**多智能体协作**和**优化的提示策略**,将单一的OpenCode助手升级为一个"**AI开发团队**"。其主要特性和最佳用法包括: - **多代理编排系统:** Sisyphus采用前沿的**多代理协同架构**,将开发任务拆解后分配给不同专长的子代理协作完成[\[79\]](https://post.smzdm.com/p/aog8xq7m#:~:text=%E4%B8%80%E4%BA%BA%E6%8A%B5%E4%B8%80%E4%B8%AA%E5%BC%80%E5%8F%91%E5%9B%A2%E9%98%9F%E7%9A%84AI%E7%BC%96%E7%A8%8B%E7%A5%9E%E5%99%A8%E5%AE%8C%E5%85%A8%E6%8C%87%E5%8D%97%20,%E5%9C%A8%E6%88%90%E6%9C%AC%E6%8E%A7%E5%88%B6%E6%96%B9%E9%9D%A2%EF%BC%8C%E8%AF%A5%E7%B3%BB%E7%BB%9F%E6%94%AF%E6%8C%81%E5%A4%9A%E6%A8%A1%E5%9E%8B%E6%B7%B7%E5%90%88%E4%BD%BF%E7%94%A8%E7%AD%96%E7%95%A5%E3%80%82%E9%80%9A%E8%BF%87%E9%85%8D%E7%BD%AE%EF%BC%8C%E5%8F%AF%E4%BB%A5%E5%B0%86)[\[41\]](https://www.cnblogs.com/treasury-manager/p/19217990#:~:text=OpenCode%20%E4%B8%80%E4%B8%AA%E7%A5%9E%E5%A5%87%E7%9A%84CLI%20%EF%BC%88%E5%8F%AF%E4%BB%A5%E8%9E%8D%E5%90%88Claude%20Code%2C%20iFLow,code%29%20%E2%86%90%20%E8%87%AA%E5%8A%A8%E4%BD%BF%E7%94%A8oracle%20%E7%9A%84%E6%A8%A1%E5%9E%8B%E2%86%93%20%E8%B0%83%E7%94%A8%40explore)。在OMO中,OpenCode不再只有一个AI在工作,而是一个主代理统筹,底下有如"Oracle预言家"、"Explorer探索者"、"Coder编码员"等多个子Agent各司其职[\[80\]](https://www.cnblogs.com/treasury-manager/p/19217990#:~:text=OpenCode%20%E4%B8%80%E4%B8%AA%E7%A5%9E%E5%A5%87%E7%9A%84CLI%20%EF%BC%88%E5%8F%AF%E4%BB%A5%E8%9E%8D%E5%90%88Claude%20Code%2C%20iFLow,code%29%20%E2%86%90%20%E8%87%AA%E5%8A%A8%E4%BD%BF%E7%94%A8oracle%20%E7%9A%84%E6%A8%A1%E5%9E%8B%E2%86%93%20%E8%B0%83%E7%94%A8%40explore)。例如,当你抛给系统一个大型需求,主代理会调用Oracle子代理来分析需求/查询知识,再调用Explorer代理搜索代码库甚至外部资料,最后由Coder代理生成代码方案,主代理综合各方反馈给出最终实现[\[80\]](https://www.cnblogs.com/treasury-manager/p/19217990#:~:text=OpenCode%20%E4%B8%80%E4%B8%AA%E7%A5%9E%E5%A5%87%E7%9A%84CLI%20%EF%BC%88%E5%8F%AF%E4%BB%A5%E8%9E%8D%E5%90%88Claude%20Code%2C%20iFLow,code%29%20%E2%86%90%20%E8%87%AA%E5%8A%A8%E4%BD%BF%E7%94%A8oracle%20%E7%9A%84%E6%A8%A1%E5%9E%8B%E2%86%93%20%E8%B0%83%E7%94%A8%40explore)。这种**软并行**的流程让复杂任务的处理效率和质量显著提高--相当于同时拥有多位AI专家在协同开发。 - **多模型混合策略:** OMO支持配置不同的LLM模型给不同子代理,以充分利用各模型所长并控制成本[\[79\]](https://post.smzdm.com/p/aog8xq7m#:~:text=%E4%B8%80%E4%BA%BA%E6%8A%B5%E4%B8%80%E4%B8%AA%E5%BC%80%E5%8F%91%E5%9B%A2%E9%98%9F%E7%9A%84AI%E7%BC%96%E7%A8%8B%E7%A5%9E%E5%99%A8%E5%AE%8C%E5%85%A8%E6%8C%87%E5%8D%97%20,%E5%9C%A8%E6%88%90%E6%9C%AC%E6%8E%A7%E5%88%B6%E6%96%B9%E9%9D%A2%EF%BC%8C%E8%AF%A5%E7%B3%BB%E7%BB%9F%E6%94%AF%E6%8C%81%E5%A4%9A%E6%A8%A1%E5%9E%8B%E6%B7%B7%E5%90%88%E4%BD%BF%E7%94%A8%E7%AD%96%E7%95%A5%E3%80%82%E9%80%9A%E8%BF%87%E9%85%8D%E7%BD%AE%EF%BC%8C%E5%8F%AF%E4%BB%A5%E5%B0%86)。例如主代理和编码代理用Claude的高性能模型,而探索/搜索代理用开源模型或较廉价的API,以降低开销[\[79\]](https://post.smzdm.com/p/aog8xq7m#:~:text=%E4%B8%80%E4%BA%BA%E6%8A%B5%E4%B8%80%E4%B8%AA%E5%BC%80%E5%8F%91%E5%9B%A2%E9%98%9F%E7%9A%84AI%E7%BC%96%E7%A8%8B%E7%A5%9E%E5%99%A8%E5%AE%8C%E5%85%A8%E6%8C%87%E5%8D%97%20,%E5%9C%A8%E6%88%90%E6%9C%AC%E6%8E%A7%E5%88%B6%E6%96%B9%E9%9D%A2%EF%BC%8C%E8%AF%A5%E7%B3%BB%E7%BB%9F%E6%94%AF%E6%8C%81%E5%A4%9A%E6%A8%A1%E5%9E%8B%E6%B7%B7%E5%90%88%E4%BD%BF%E7%94%A8%E7%AD%96%E7%95%A5%E3%80%82%E9%80%9A%E8%BF%87%E9%85%8D%E7%BD%AE%EF%BC%8C%E5%8F%AF%E4%BB%A5%E5%B0%86)。有文章指出OMO可以通过配置,实现"大模型做决策,小模型跑腿"的组合,在保证效果的同时节约预算[\[79\]](https://post.smzdm.com/p/aog8xq7m#:~:text=%E4%B8%80%E4%BA%BA%E6%8A%B5%E4%B8%80%E4%B8%AA%E5%BC%80%E5%8F%91%E5%9B%A2%E9%98%9F%E7%9A%84AI%E7%BC%96%E7%A8%8B%E7%A5%9E%E5%99%A8%E5%AE%8C%E5%85%A8%E6%8C%87%E5%8D%97%20,%E5%9C%A8%E6%88%90%E6%9C%AC%E6%8E%A7%E5%88%B6%E6%96%B9%E9%9D%A2%EF%BC%8C%E8%AF%A5%E7%B3%BB%E7%BB%9F%E6%94%AF%E6%8C%81%E5%A4%9A%E6%A8%A1%E5%9E%8B%E6%B7%B7%E5%90%88%E4%BD%BF%E7%94%A8%E7%AD%96%E7%95%A5%E3%80%82%E9%80%9A%E8%BF%87%E9%85%8D%E7%BD%AE%EF%BC%8C%E5%8F%AF%E4%BB%A5%E5%B0%86)。这对需要长时间运行的自动化Agent特别重要。实际使用中,很多人将GPT-4、Claude、甚至本地模型共同挂接到OMO,让它们各尽其用,整套系统智能且经济。 - **全流程自动执行(Ultrawork模式):** **"不用读文档,上来就干!"** 这是不少OMO用户的直观感受[\[81\]](https://github.com/code-yeongyu/oh-my-opencode/blob/dev/README.zh-cn.md#:~:text=oh,%E8%80%81%E5%AE%9E%E8%AF%B4%EF%BC%8C%E7%94%9A%E8%87%B3%E4%B8%8D%E7%94%A8%E8%B4%B9%E5%BF%83%E8%AF%BB%E6%96%87%E6%A1%A3%E3%80%82%E5%8F%AA%E9%9C%80%E5%86%99%E4%BD%A0%E7%9A%84%E6%8F%90%E7%A4%BA%E3%80%82%E5%8C%85%E5%90%AB%27ultrawork%27%20%E5%85%B3%E9%94%AE%E8%AF%8D%E3%80%82Sisyphus%20%E4%BC%9A%E5%88%86%E6%9E%90%E7%BB%93%E6%9E%84%EF%BC%8C%E6%94%B6%E9%9B%86%E4%B8%8A%E4%B8%8B%E6%96%87%EF%BC%8C%E6%8C%96%E6%8E%98%E5%A4%96%E9%83%A8%E6%BA%90%E4%BB%A3%E7%A0%81%EF%BC%8C%E7%84%B6%E5%90%8E%E6%8C%81%E7%BB%AD%E6%8E%A8%E8%BF%9B)。OMO内置了一种特殊的提示触发机制,例如当在对话中包含关键词"ultrawork"时,Sisyphus会**自动接管任务**,从分析、检索到编码一步步执行下去,几乎无需人介入[\[81\]](https://github.com/code-yeongyu/oh-my-opencode/blob/dev/README.zh-cn.md#:~:text=oh,%E8%80%81%E5%AE%9E%E8%AF%B4%EF%BC%8C%E7%94%9A%E8%87%B3%E4%B8%8D%E7%94%A8%E8%B4%B9%E5%BF%83%E8%AF%BB%E6%96%87%E6%A1%A3%E3%80%82%E5%8F%AA%E9%9C%80%E5%86%99%E4%BD%A0%E7%9A%84%E6%8F%90%E7%A4%BA%E3%80%82%E5%8C%85%E5%90%AB%27ultrawork%27%20%E5%85%B3%E9%94%AE%E8%AF%8D%E3%80%82Sisyphus%20%E4%BC%9A%E5%88%86%E6%9E%90%E7%BB%93%E6%9E%84%EF%BC%8C%E6%94%B6%E9%9B%86%E4%B8%8A%E4%B8%8B%E6%96%87%EF%BC%8C%E6%8C%96%E6%8E%98%E5%A4%96%E9%83%A8%E6%BA%90%E4%BB%A3%E7%A0%81%EF%BC%8C%E7%84%B6%E5%90%8E%E6%8C%81%E7%BB%AD%E6%8E%A8%E8%BF%9B)。开发者只要提出高层次目标,如"ultrawork: 实现一个博客网站的前后端,包括用户注册、发帖、评论"等,接下来Sisyphus就会自主规划子任务、调用子代理查资料、写代码、测试,迭代直到完成[\[81\]](https://github.com/code-yeongyu/oh-my-opencode/blob/dev/README.zh-cn.md#:~:text=oh,%E8%80%81%E5%AE%9E%E8%AF%B4%EF%BC%8C%E7%94%9A%E8%87%B3%E4%B8%8D%E7%94%A8%E8%B4%B9%E5%BF%83%E8%AF%BB%E6%96%87%E6%A1%A3%E3%80%82%E5%8F%AA%E9%9C%80%E5%86%99%E4%BD%A0%E7%9A%84%E6%8F%90%E7%A4%BA%E3%80%82%E5%8C%85%E5%90%AB%27ultrawork%27%20%E5%85%B3%E9%94%AE%E8%AF%8D%E3%80%82Sisyphus%20%E4%BC%9A%E5%88%86%E6%9E%90%E7%BB%93%E6%9E%84%EF%BC%8C%E6%94%B6%E9%9B%86%E4%B8%8A%E4%B8%8B%E6%96%87%EF%BC%8C%E6%8C%96%E6%8E%98%E5%A4%96%E9%83%A8%E6%BA%90%E4%BB%A3%E7%A0%81%EF%BC%8C%E7%84%B6%E5%90%8E%E6%8C%81%E7%BB%AD%E6%8E%A8%E8%BF%9B)。在B站等平台的演示视频中,可以看到OpenCode + OMO真的实现了**全程零干预**地产出项目雏形[\[82\]](https://www.youtube.com/watch?v=twFjLiy2Pmc#:~:text=%E8%A7%86%E9%A2%91%E7%AE%80%E4%BB%8B%EF%BC%9A%20%E6%9C%AC%E6%9C%9F%E8%A7%86%E9%A2%91%E8%AF%A6%E7%BB%86%E6%BC%94%E7%A4%BA%E4%BA%86%E5%A6%82%E4%BD%95%E5%9C%A8Opencode%E4%B8%AD%E4%BD%BF%E7%94%A8%E6%9C%80%E5%BC%BA%E5%BC%80%E6%BA%90%E6%8F%92%E4%BB%B6Oh%20My%20Opencode%EF%BC%88OMO%EF%BC%89%EF%BC%8C%E5%B0%86%E5%8D%95%E4%B8%80AI%E7%BC%96%E7%A8%8B%E5%8A%A9%E6%89%8B%E5%8D%87%E7%BA%A7%E4%B8%BA%E5%A4%9A%E6%A8%A1%E5%9E%8B%E5%8D%8F%E4%BD%9C%E7%9A%84AI%E5%BC%80%E5%8F%91%E5%9B%A2%E9%98%9F%EF%BC%81)[\[83\]](https://x.com/AISuperDomain/status/2009823408301994209#:~:text=OpenCode%E7%9A%84%E6%9C%80%E5%BC%BA%E5%BC%80%E6%BA%90%E6%8F%92%E4%BB%B6oh%20my%20opencode%E7%A1%AE%E5%AE%9E%E9%9D%9E%E5%B8%B8%E5%A5%BD%E7%94%A8%20%E8%A7%86%E9%A2%91%E7%AE%80%E4%BB%8B%EF%BC%9A%20%E6%9C%AC%E6%9C%9F%E8%A7%86%E9%A2%91%E8%AF%A6%E7%BB%86%E6%BC%94%E7%A4%BA%E4%BA%86%E5%A6%82%E4%BD%95%E5%9C%A8Opencode%E4%B8%AD%E4%BD%BF%E7%94%A8%E6%9C%80%E5%BC%BA%E5%BC%80%E6%BA%90%E6%8F%92%E4%BB%B6Oh,My%20Opencode%EF%BC%88OMO%EF%BC%89%EF%BC%8C%E5%B0%86%E5%8D%95%E4%B8%80AI%E7%BC%96%E7%A8%8B%E5%8A%A9%E6%89%8B%E5%8D%87%E7%BA%A7%E4%B8%BA%E5%A4%9A%E6%A8%A1%E5%9E%8B%E5%8D%8F%E4%BD%9C%E7%9A%84AI%E5%BC%80%E5%8F%91%E5%9B%A2%E9%98%9F%EF%BC%81%20%E6%A0%B8%E5%BF%83%E5%86%85%E5%AE%B9%EF%BC%9AOMO%E6%8F%92%E4%BB%B6)。当然如此长链路的执行可能偶有偏差,但这展示了Agentic AI开发的巨大潜力。对于使用OMO的我们,建议在较明确且可分解的项目上尝试"Ultrawork"模式,静待AI团队跑完全流程,再对结果集中验收修改。 - **可定制 Hooks 与插件:** OMO不仅提供默认的智能体团队,还允许高级用户扩展自定义**Hooks**和插件逻辑[\[84\]](https://blog.csdn.net/gitblog_00895/article/details/144862506#:~:text=oh,)。其Hooks系统支持在AI执行过程的关键节点插入自定义代码,以实现特殊监控或操作[\[84\]](https://blog.csdn.net/gitblog_00895/article/details/144862506#:~:text=oh,)。比如可以在每次子代理调用外部API前触发Hook检查额度,或在生成代码后通过Hook自动运行格式化工具[\[84\]](https://blog.csdn.net/gitblog_00895/article/details/144862506#:~:text=oh,)。这赋予OMO极高的可扩展性,能够适应各种个性化需求。因此最佳实践是:在熟练掌握默认OMO能力后,再根据自己团队的流程编写Hooks或定制代理,打造**专属的AI流水线**。目前社区也涌现了许多OMO的附加插件和配置心得,可借鉴来强化你的AI团队。 作为OpenCode的"超充电"版,Oh My OpenCode显著提升了AI编程助手的自主性和专业度。有用户戏称:"用了OMO,仿佛一个人带着一支AI舰队写代码"--一人顶一个团队的效率令人震撼[\[79\]](https://post.smzdm.com/p/aog8xq7m#:~:text=%E4%B8%80%E4%BA%BA%E6%8A%B5%E4%B8%80%E4%B8%AA%E5%BC%80%E5%8F%91%E5%9B%A2%E9%98%9F%E7%9A%84AI%E7%BC%96%E7%A8%8B%E7%A5%9E%E5%99%A8%E5%AE%8C%E5%85%A8%E6%8C%87%E5%8D%97%20,%E5%9C%A8%E6%88%90%E6%9C%AC%E6%8E%A7%E5%88%B6%E6%96%B9%E9%9D%A2%EF%BC%8C%E8%AF%A5%E7%B3%BB%E7%BB%9F%E6%94%AF%E6%8C%81%E5%A4%9A%E6%A8%A1%E5%9E%8B%E6%B7%B7%E5%90%88%E4%BD%BF%E7%94%A8%E7%AD%96%E7%95%A5%E3%80%82%E9%80%9A%E8%BF%87%E9%85%8D%E7%BD%AE%EF%BC%8C%E5%8F%AF%E4%BB%A5%E5%B0%86)[\[85\]](https://post.smzdm.com/p/aog8xq7m#:~:text=%E5%9C%A8%E6%88%90%E6%9C%AC%E6%8E%A7%E5%88%B6%E6%96%B9%E9%9D%A2%EF%BC%8C%E8%AF%A5%E7%B3%BB%E7%BB%9F%E6%94%AF%E6%8C%81%E5%A4%9A%E6%A8%A1%E5%9E%8B%E6%B7%B7%E5%90%88%E4%BD%BF%E7%94%A8%E7%AD%96%E7%95%A5%E3%80%82%E9%80%9A%E8%BF%87%E9%85%8D%E7%BD%AE%EF%BC%8C%E5%8F%AF%E4%BB%A5%E5%B0%86%20)。不过也需注意,OMO毕竟处于新兴阶段,全自动模式下如果任务不清晰,AI可能产生偏差 output。因此,**明确的任务描述**依然是成功的前提。在对OMO下达指令时,尽量描述清楚最终期望和约束条件,并监控关键输出节点。如果结果不理想,可以通过交互引导主代理调整方案或重试特定子任务。相信随着社区不断优化Sisyphus的提示技巧和反馈机制,OMO会变得越来越"稳"。对于敢于尝新的开发者,它无疑是值得深入研究的利器:熟练掌握后,你将拥有前所未有的**开发自动化能力**,显著缩短从想法到产品的路径。 ## 总结 从Claude Code Skills到OpenCode再到Oh My OpenCode,我们见证了AI辅助开发从**手工提示**走向**模块化复用**、从**单点对话**走向**多智能体协作**的演进。Claude Code Skills教会我们如何将**提示词工程升级为可复用的技能模块**--通过YAML结构化知识,渐进披露上下文,让AI随时调用专业技能包,大幅减少重复劳动[\[5\]](https://cloud.tencent.com/developer/article/2616585#:~:text=Claude%20Skills%E6%98%AFAnthropic%E6%8E%A8%E5%87%BA%E7%9A%84AI%E5%8A%9F%E8%83%BD%E9%9D%A9%E5%91%BD%EF%BC%8C%E5%8F%AF%E5%B0%86%E7%94%A8%E6%88%B7%E4%BD%BF%E7%94%A8AI%E7%9A%84%E4%B9%A0%E6%83%AF%E6%96%87%E4%BB%B6%E5%8C%96%E7%AE%A1%E7%90%86%E3%80%82%E5%AE%83%E8%83%BD%E8%A7%A3%E5%86%B3Claude%E5%81%A5%E5%BF%98%E3%80%81%E9%9C%80%E9%87%8D%E5%A4%8D%E6%8F%90%E7%A4%BA%E8%AF%8D%E7%9A%84%E9%97%AE%E9%A2%98%EF%BC%8C%E5%B0%86%E4%BB%BB%E5%8A%A1%E8%AF%B4%E6%98%8E%E3%80%81%E5%B7%A5%E5%85%B7%E4%BB%A3%E7%A0%81%E7%AD%89%E6%89%93%E5%8C%85%E6%88%90%20)[\[6\]](https://www.facebook.com/iamvista/photos/%E6%9F%90%E5%A4%A9%E6%B7%B1%E5%A4%9C%E6%88%91%E6%AD%A3%E5%9C%A8%E8%B6%95%E4%B8%80%E4%BB%BD%E6%96%87%E4%BB%B6%E5%A4%A9%E5%95%8A%E5%90%8C%E6%A8%A3%E7%9A%84%E6%9E%B6%E6%A7%8B%E5%90%8C%E6%A8%A3%E7%9A%84%E8%AA%9E%E6%B0%A3%E5%90%8C%E6%A8%A3%E7%9A%84%E6%A0%BC%E5%BC%8F%E8%A6%81%E6%B1%82%E4%BD%86%E6%88%91%E5%8F%88%E5%BE%97%E9%87%8D%E6%96%B0%E6%89%93%E4%B8%80%E6%AC%A1%E8%AB%8B%E7%94%A8%E9%80%99%E5%80%8B%E6%A0%BC%E5%BC%8F%E8%AB%8B%E5%85%88%E5%95%8F%E4%B8%89%E5%80%8B%E6%BE%84%E6%B8%85%E5%95%8F%E9%A1%8C%E8%AB%8B%E6%8A%8A%E8%BC%B8%E5%87%BA%E5%88%86%E6%88%90%E5%9B%9B%E6%AE%B5%E8%AB%8B%E9%99%84%E4%B8%8A%E5%8F%AF%E7%9B%B4%E6%8E%A5%E8%B2%BC%E5%88%B0-notion-%E7%9A%84/10162293093624053/#:~:text=,%E6%88%91%E6%80%8E%E9%BA%BC%E5%81%9A%E4%B8%80%E5%80%8B%E5%A5%BDskill%EF%BC%8C%E8%AE%8A%E6%88%90skill)。OpenCode则以开源之力,将AI编程助手引入开发者日常工具链,在终端或IDE中实现了与AI无缝结对编程[\[48\]](https://zhuanlan.zhihu.com/p/1991170184573122515#:~:text=OpenCode%20%E6%98%AF%E4%B8%80%E4%B8%AAAI%20%E7%BC%96%E7%A8%8B%E5%8A%A9%E6%89%8B%EF%BC%8C%E8%B7%91%E5%9C%A8%E4%BD%A0%E7%9A%84%E7%BB%88%E7%AB%AF%E9%87%8C%EF%BC%88%E5%B0%B1%E6%98%AF%E9%82%A3%E4%B8%AA%E9%BB%91%E8%89%B2%E7%AA%97%E5%8F%A3%EF%BC%89%E3%80%82%20%E4%BD%A0%E8%B7%9F%E5%AE%83%E8%AF%B4%E8%AF%9D%EF%BC%8C%E5%AE%83%E5%B0%B1%E5%B8%AE%E4%BD%A0%E5%86%99%E4%BB%A3%E7%A0%81%E3%80%82%20,%E2%86%92%20%E5%AE%83%E6%94%B9)。其Plan/Build双模式和Undo/Redo等机制,被证明是安全高效地让AI参与编码的最佳实践[\[63\]](https://opencode.ai/docs#:~:text=1)[\[71\]](https://opencode.ai/docs#:~:text=Let%E2%80%99s%20say%20you%20ask%20OpenCode,to%20make%20some%20changes)。而Oh My OpenCode更进一步,引入多Agent架构和自动化工作流,展示了"AI团队开发"的雏形[\[79\]](https://post.smzdm.com/p/aog8xq7m#:~:text=%E4%B8%80%E4%BA%BA%E6%8A%B5%E4%B8%80%E4%B8%AA%E5%BC%80%E5%8F%91%E5%9B%A2%E9%98%9F%E7%9A%84AI%E7%BC%96%E7%A8%8B%E7%A5%9E%E5%99%A8%E5%AE%8C%E5%85%A8%E6%8C%87%E5%8D%97%20,%E5%9C%A8%E6%88%90%E6%9C%AC%E6%8E%A7%E5%88%B6%E6%96%B9%E9%9D%A2%EF%BC%8C%E8%AF%A5%E7%B3%BB%E7%BB%9F%E6%94%AF%E6%8C%81%E5%A4%9A%E6%A8%A1%E5%9E%8B%E6%B7%B7%E5%90%88%E4%BD%BF%E7%94%A8%E7%AD%96%E7%95%A5%E3%80%82%E9%80%9A%E8%BF%87%E9%85%8D%E7%BD%AE%EF%BC%8C%E5%8F%AF%E4%BB%A5%E5%B0%86)[\[85\]](https://post.smzdm.com/p/aog8xq7m#:~:text=%E5%9C%A8%E6%88%90%E6%9C%AC%E6%8E%A7%E5%88%B6%E6%96%B9%E9%9D%A2%EF%BC%8C%E8%AF%A5%E7%B3%BB%E7%BB%9F%E6%94%AF%E6%8C%81%E5%A4%9A%E6%A8%A1%E5%9E%8B%E6%B7%B7%E5%90%88%E4%BD%BF%E7%94%A8%E7%AD%96%E7%95%A5%E3%80%82%E9%80%9A%E8%BF%87%E9%85%8D%E7%BD%AE%EF%BC%8C%E5%8F%AF%E4%BB%A5%E5%B0%86%20)。利用OMO,复杂项目也许只需提出愿景,AI代理即可协同完成大部分实现,开发者转为高层监督和收尾调整的角色。 当然,再强大的AI工具也需要人来驾驭。**明确的问题定义、合理的分工接口、及时的反馈干预**,依然是成功应用AI的关键。对于Golang+Vue3这类前后端结合的项目,我们强调了提供充足上下文、分步指导AI、结合测试验证等提示词设计要点,以充分发挥Claude/GPT-4的长处[\[39\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=match%20at%20L316%20Give%20Claude,context%2C%20the%20better%20the%20suggestions)[\[40\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=match%20at%20L322%20When%20Claude,things%20you%E2%80%99ve%20already%20decided%20against)。当这些经验与像OpenCode这样的平台结合,AI辅助编程就真正融入了软件开发生命周期。从代码生成、重构到调试、测试,各环节皆有AI参与,其价值不再停留在"生成几段代码"上,而是成为一种**开发范式的升级**。 正如Anthropic在文档中所言:未来的开发者将不仅仅会写代码,更要善于**打造和利用AI技能**[\[23\]](https://claudecn.com/blog/claude-skills-architecture/#:~:text=1)。通过模块化的Skills,我们可以将个人经验沉淀,让AI持续进化成为领域专家;通过工具与插件,我们可以让AI深度融入团队协作,实现人机共创。当下的这些探索,正是迈向"自我进化软件"的垫脚石。希望本指南的调研和总结,能帮助你在实践中少走弯路,加速拥抱AI赋能的全新开发模式。在不远的将来,写代码或许就像与聪明的同事对话一样自然,而我们将有更多时间专注于创造真正有价值的东西。让我们拭目以待,也积极参与这一场AI与编程的革命吧! **参考文献:** - Anthropic Claude Skills 官方文档及教程[\[9\]](https://ide.unitmesh.cc/spec/claude-skill#:~:text=%E6%8A%80%E8%83%BD%E5%AE%9A%E4%B9%89%E6%A0%BC%E5%BC%8F)[\[86\]](https://claudecn.com/blog/claude-skills-architecture/#:~:text=%E4%B8%BA%E4%BA%86%E9%81%BF%E5%85%8D%E2%80%9C%E4%BB%8B%E7%BB%8D%E5%8A%9F%E8%83%BD%E4%BD%86%E7%BC%BA%E5%B0%91%E5%8F%AF%E6%A3%80%E9%AA%8C%E7%BB%93%E8%AE%BA%E2%80%9D%EF%BC%8C%E6%9C%AC%E6%96%87%E5%85%88%E6%8A%8A%20Skills%20%E7%9A%84%E6%9E%B6%E6%9E%84%E4%BB%B7%E5%80%BC%E6%94%B6%E6%95%9B%E6%88%90%E4%B8%80%E5%8F%A5%E5%8F%AF%E9%AA%8C%E8%AF%81%E7%9A%84%E5%91%BD%E9%A2%98%EF%BC%9A) - 社区博客与实践案例(CSDN、知乎等)对Claude Code/Skills的解读[\[15\]](https://blog.csdn.net/yangshangwei/article/details/156836796#:~:text=%E4%B8%80%E4%B8%AA%E5%8F%AF%E8%90%BD%E5%9C%B0%E7%9A%84%E4%B8%89%E6%AD%A5%E6%B3%95,3%20%E7%AC%AC%E4%B8%89%E6%AD%A5%EF%BC%9A%E6%8A%8AClaude%20%E5%BD%93%E2%80%9C%E6%90%AD%E5%AD%90%E2%80%9D%EF%BC%8C%E8%80%8C%E4%B8%8D%E6%98%AF%E6%90%9C%E7%B4%A2%E6%A1%86)[\[18\]](https://hbwdj.gov.cn/appbdetail-imqqsmrp9897358.d#:~:text=%E9%AA%97%E4%BD%A0%E7%9A%84%EF%BC%8C%E5%85%B6%E5%AE%9EAI%E6%A0%B9%E6%9C%AC%E4%B8%8D%E9%9C%80%E8%A6%81%E9%82%A3%E4%B9%88%E5%A4%9A%E6%8F%90%E7%A4%BA%E8%AF%8D%E3%80%82%20%E4%BD%A0%E5%8F%AA%E9%9C%80%E8%A6%81%E8%B0%83%E7%94%A8AI%20%E6%9C%AC%E8%BA%AB%E7%9A%84%E2%80%9CSkill%20Creator%E2%80%9D%E6%8A%80%E8%83%BD%EF%BC%8C%E7%94%A8%E4%BD%A0%E7%9A%84%E8%AF%AD%E8%A8%80%E6%8F%8F%E8%BF%B0%E8%87%AA%E5%B7%B1%E7%9A%84%E9%9C%80%E6%B1%82%EF%BC%8C%E8%AE%A9AI%E8%87%AA%E5%8A%A8%E5%B8%AE%E4%BD%A0%E7%94%9F%E6%88%90%E4%B8%80%E9%97%A8%E6%8A%80%E8%83%BD%EF%BC%8C%E4%BD%BF%E7%94%A8%E8%B5%B7%E6%9D%A5%E9%9D%9E%E5%B8%B8%E5%8F%8B%E5%A5%BD%EF%BC%8CAI%E4%BC%9A%E4%B8%80%E6%AD%A5%E6%AD%A5%E5%BC%95%E5%AF%BC%E4%BD%A0%E8%AF%B4%E5%87%BA%E4%BD%A0%E7%9A%84%E9%9C%80%E6%B1%82%EF%BC%8C%E4%BD%A0%E5%8F%AA%20) - OpenCode项目仓库README及官方文档[\[51\]](https://github.com/anomalyco/opencode#:~:text=OpenCode%20includes%20two%20built,key)[\[52\]](https://opencode.ai/docs#:~:text=1) - Oh My OpenCode项目介绍、社区经验分享[\[80\]](https://www.cnblogs.com/treasury-manager/p/19217990#:~:text=OpenCode%20%E4%B8%80%E4%B8%AA%E7%A5%9E%E5%A5%87%E7%9A%84CLI%20%EF%BC%88%E5%8F%AF%E4%BB%A5%E8%9E%8D%E5%90%88Claude%20Code%2C%20iFLow,code%29%20%E2%86%90%20%E8%87%AA%E5%8A%A8%E4%BD%BF%E7%94%A8oracle%20%E7%9A%84%E6%A8%A1%E5%9E%8B%E2%86%93%20%E8%B0%83%E7%94%A8%40explore)[\[81\]](https://github.com/code-yeongyu/oh-my-opencode/blob/dev/README.zh-cn.md#:~:text=oh,%E8%80%81%E5%AE%9E%E8%AF%B4%EF%BC%8C%E7%94%9A%E8%87%B3%E4%B8%8D%E7%94%A8%E8%B4%B9%E5%BF%83%E8%AF%BB%E6%96%87%E6%A1%A3%E3%80%82%E5%8F%AA%E9%9C%80%E5%86%99%E4%BD%A0%E7%9A%84%E6%8F%90%E7%A4%BA%E3%80%82%E5%8C%85%E5%90%AB%27ultrawork%27%20%E5%85%B3%E9%94%AE%E8%AF%8D%E3%80%82Sisyphus%20%E4%BC%9A%E5%88%86%E6%9E%90%E7%BB%93%E6%9E%84%EF%BC%8C%E6%94%B6%E9%9B%86%E4%B8%8A%E4%B8%8B%E6%96%87%EF%BC%8C%E6%8C%96%E6%8E%98%E5%A4%96%E9%83%A8%E6%BA%90%E4%BB%A3%E7%A0%81%EF%BC%8C%E7%84%B6%E5%90%8E%E6%8C%81%E7%BB%AD%E6%8E%A8%E8%BF%9B) - Korbinian Schleifer: _Learning Go and Vue with Claude AI as my Pair Programmer_[\[39\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=match%20at%20L316%20Give%20Claude,context%2C%20the%20better%20the%20suggestions)[\[40\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=match%20at%20L322%20When%20Claude,things%20you%E2%80%99ve%20already%20decided%20against) - 更多详见上述来源引用[\[8\]](https://ide.unitmesh.cc/spec/claude-skill#:~:text=my,%E8%BE%85%E5%8A%A9%E8%84%9A%E6%9C%AC%EF%BC%88%E5%8F%AF%E9%80%89%EF%BC%89)[\[68\]](https://opencode.ai/docs#:~:text=3)等。 [\[1\]](https://ide.unitmesh.cc/spec/claude-skill#:~:text=%E4%BB%80%E4%B9%88%E6%98%AF%20Claude%20Skills%EF%BC%9F) [\[7\]](https://ide.unitmesh.cc/spec/claude-skill#:~:text=%E6%8A%80%E8%83%BD%E7%BB%93%E6%9E%84%EF%BC%9A) [\[8\]](https://ide.unitmesh.cc/spec/claude-skill#:~:text=my,%E8%BE%85%E5%8A%A9%E8%84%9A%E6%9C%AC%EF%BC%88%E5%8F%AF%E9%80%89%EF%BC%89) [\[9\]](https://ide.unitmesh.cc/spec/claude-skill#:~:text=%E6%8A%80%E8%83%BD%E5%AE%9A%E4%B9%89%E6%A0%BC%E5%BC%8F) [\[10\]](https://ide.unitmesh.cc/spec/claude-skill#:~:text=,%E7%B1%BB%E5%88%AB2) [\[11\]](https://ide.unitmesh.cc/spec/claude-skill#:~:text=) [\[12\]](https://ide.unitmesh.cc/spec/claude-skill#:~:text=Anthropic%20Claude%20Skills%20%E6%98%AF%E5%8C%85%E5%90%AB%E6%8C%87%E4%BB%A4%E3%80%81%E8%84%9A%E6%9C%AC%E5%92%8C%E8%B5%84%E6%BA%90%E7%9A%84%E6%96%87%E4%BB%B6%E5%A4%B9%EF%BC%8CClaude%20%E5%8A%A8%E6%80%81%E5%8A%A0%E8%BD%BD%E4%BB%A5%E6%8F%90%E9%AB%98%E4%B8%93%E4%B8%9A%E4%BB%BB%E5%8A%A1%E7%9A%84%E6%80%A7%E8%83%BD%E3%80%82) [\[13\]](https://ide.unitmesh.cc/spec/claude-skill#:~:text=) [\[16\]](https://ide.unitmesh.cc/spec/claude-skill#:~:text=,%E9%A2%84%E7%AE%97%E6%98%8E%E7%BB%86%20%E5%90%AF%E7%94%A8%E4%BF%AE%E8%AE%A2%E8%B7%9F%E8%B8%AA%E4%BB%A5%E4%BE%9B%E5%AE%A1%E6%9F%A5) [\[17\]](https://ide.unitmesh.cc/spec/claude-skill#:~:text=) Claude Skill | AutoDev - Tailor Your AI Coding Experience [\[2\]](https://www.bilibili.com/opus/1132124852115734530#:~:text=Claude%20Skills%20%E6%9C%AC%E8%B4%A8%E4%B8%8A%E6%98%AF%E4%B8%80%E7%A7%8D,) [\[30\]](https://www.bilibili.com/opus/1132124852115734530#:~:text=Claude%20Skills%20%E5%AE%9E%E6%88%98%E6%8C%87%E5%8D%97%EF%BC%9A3%20%E5%88%86%E9%92%9F%E6%90%9E%E5%AE%9APPT%E3%80%81%E6%B5%B7%E6%8A%A5%E4%B8%8ELogo%20,%E5%AE%83%E5%B0%86%E5%B8%B8%E8%A7%81%E7%9A%84%E5%8A%9E%E5%85%AC%E4%BB%BB%E5%8A%A1%28%E5%A6%82Excel%20%E6%95%B0%E6%8D%AE%E5%88%86%E6%9E%90%E3%80%81PPT%20%E6%BC%94%E7%A4%BA%E7%94%9F%E6%88%90%E3%80%81%E6%96%87%E6%A1%A3%E5%A4%84%E7%90%86%E3%80%81%E5%93%81%E7%89%8C%E8%AE%BE%E8%AE%A1%E7%AD%89%29%E5%B0%81%E8%A3%85%E6%88%90%E5%8F%AF%E5%A4%8D%E7%94%A8%E7%9A%84%E6%8A%80%E8%83%BD%E6%A8%A1%E5%9D%97%E3%80%82%E8%BF%99%E7%A7%8D%E8%AE%BE%E8%AE%A1%E7%90%86%E5%BF%B5%E7%9A%84%E6%A0%B8%E5%BF%83%E4%BC%98%E5%8A%BF) Claude Skills 实战指南:3 分钟搞定PPT、海报与Logo - Bilibili [\[3\]](https://claudecn.com/#:~:text=Agent%20Skills%20%E5%8F%AF%E5%A4%8D%E7%94%A8%E7%9A%84%E6%8A%80%E8%83%BD%E7%B3%BB%E7%BB%9F%EF%BC%8C%E8%AE%A9%20Claude%20%E6%8E%8C%E6%8F%A1%E7%89%B9%E5%AE%9A%E9%A2%86%E5%9F%9F%E4%B8%93%E4%B8%9A%E7%9F%A5%E8%AF%86%E3%80%82%E5%8C%85%E6%8B%AC,%E8%BF%9E%E6%8E%A5%E5%A4%96%E9%83%A8%E5%B7%A5%E5%85%B7%E5%92%8C%E6%95%B0%E6%8D%AE%E6%BA%90%EF%BC%8C%E6%89%A9%E5%B1%95%20Claude%20%E7%9A%84%E8%83%BD%E5%8A%9B%E8%BE%B9%E7%95%8C%EF%BC%8C%E6%9E%84%E5%BB%BA%E5%BC%BA%E5%A4%A7%E7%9A%84%20AI%20%E5%BA%94%E7%94%A8%E7%94%9F%E6%80%81%E3%80%82) Claude 中文 - Claude AI 开发技术社区 [\[4\]](https://claudecn.com/blog/claude-skills-architecture/#:~:text=,3.2.2%20Markdown%20%E6%8C%87%E4%BB%A4%E6%AD%A3%E6%96%87) [\[23\]](https://claudecn.com/blog/claude-skills-architecture/#:~:text=1) [\[24\]](https://claudecn.com/blog/claude-skills-architecture/#:~:text=%E4%B8%BA%E4%BA%86%E9%81%BF%E5%85%8D%E2%80%9C%E4%BB%8B%E7%BB%8D%E5%8A%9F%E8%83%BD%E4%BD%86%E7%BC%BA%E5%B0%91%E5%8F%AF%E6%A3%80%E9%AA%8C%E7%BB%93%E8%AE%BA%E2%80%9D%EF%BC%8C%E6%9C%AC%E6%96%87%E5%85%88%E6%8A%8A%20Skills%20%E7%9A%84%E6%9E%B6%E6%9E%84%E4%BB%B7%E5%80%BC%E6%94%B6%E6%95%9B%E6%88%90%E4%B8%80%E5%8F%A5%E5%8F%AF%E9%AA%8C%E8%AF%81%E7%9A%84%E5%91%BD%E9%A2%98%EF%BC%9A) [\[25\]](https://claudecn.com/blog/claude-skills-architecture/#:~:text=Skills%20%E6%8A%8A%20LLM%20%E7%B3%BB%E7%BB%9F%E4%BB%8E%E2%80%9C%E6%96%87%E6%9C%AC%E5%A0%86%E5%8F%A0%E7%9A%84%E5%8D%95%E4%BD%93%E6%8F%90%E7%A4%BA%E8%AF%8D%E2%80%9D%EF%BC%8C%E9%87%8D%E6%9E%84%E4%B8%BA%E2%80%9C%E5%8F%AF%E7%89%88%E6%9C%AC%E5%8C%96%E3%80%81%E5%8F%AF%E5%AE%A1%E8%AE%A1%E3%80%81%E5%8F%AF%E7%BB%84%E5%90%88%E7%9A%84%E8%BF%90%E8%A1%8C%E6%97%B6%E6%A8%A1%E5%9D%97%E2%80%9D%EF%BC%9B%E6%A0%B8%E5%BF%83%E6%94%B6%E7%9B%8A%E6%9D%A5%E8%87%AA%E4%B8%89%E4%BB%B6%E4%BA%8B%EF%BC%9A%E4%B8%8A%E4%B8%8B%E6%96%87%E9%A2%84%E7%AE%97%E5%8F%AF%E6%8E%A7%E3%80%81%E6%89%A7%E8%A1%8C%E8%B7%AF%E5%BE%84%E5%8F%AF%E6%8E%A7%E3%80%81%E6%9D%83%E9%99%90%E8%BE%B9%E7%95%8C%E5%8F%AF%E6%8E%A7%E3%80%82) [\[26\]](https://claudecn.com/blog/claude-skills-architecture/#:~:text=,3.%20%E6%8A%80%E6%9C%AF%E8%A7%A3%E6%9E%84%EF%BC%9ASkill%20%E7%9A%84%E7%89%A9%E7%90%86%E5%BD%A2%E6%80%81%E4%B8%8E%E8%A7%84%E8%8C%83) [\[27\]](https://claudecn.com/blog/claude-skills-architecture/#:~:text=,2.3%20%E6%8A%8A%E2%80%9C%E5%88%86%E5%B1%82%E6%8A%AB%E9%9C%B2%E2%80%9D%E6%8F%90%E5%8D%87%E4%B8%BA%E8%BF%90%E8%A1%8C%E6%97%B6%E7%8A%B6%E6%80%81%E6%9C%BA) [\[28\]](https://claudecn.com/blog/claude-skills-architecture/#:~:text=,%E6%9D%83%E9%99%90%E8%BE%B9%E7%95%8C%E5%8F%AF%E6%8E%A7%EF%BC%9A%E7%94%A8%E6%B2%99%E7%AE%B1%E3%80%81%E7%BD%91%E7%BB%9C%E4%BB%A3%E7%90%86%E4%B8%8E%E6%9D%83%E9%99%90%E6%8F%90%E7%A4%BA%EF%BC%8C%E6%8A%8A%E5%B7%A5%E5%85%B7%E6%89%A7%E8%A1%8C%E9%9D%A2%E6%94%B6%E6%95%9B%E5%88%B0%E5%8F%AF%E5%AE%A1%E8%AE%A1%E3%80%81%E5%8F%AF%E6%B2%BB%E7%90%86%E7%9A%84%E8%BE%B9%E7%95%8C%E5%86%85%E3%80%82) [\[29\]](https://claudecn.com/blog/claude-skills-architecture/#:~:text=%2A%204.1%20%E7%9B%91%E7%9D%A3%E8%80%85,6.2%20%E5%8D%8F%E5%90%8C%E6%9E%B6%E6%9E%84%EF%BC%9ASkill%20%E4%BD%9C%E4%B8%BA%E7%BC%96%E6%8E%92%E8%80%85) [\[86\]](https://claudecn.com/blog/claude-skills-architecture/#:~:text=%E4%B8%BA%E4%BA%86%E9%81%BF%E5%85%8D%E2%80%9C%E4%BB%8B%E7%BB%8D%E5%8A%9F%E8%83%BD%E4%BD%86%E7%BC%BA%E5%B0%91%E5%8F%AF%E6%A3%80%E9%AA%8C%E7%BB%93%E8%AE%BA%E2%80%9D%EF%BC%8C%E6%9C%AC%E6%96%87%E5%85%88%E6%8A%8A%20Skills%20%E7%9A%84%E6%9E%B6%E6%9E%84%E4%BB%B7%E5%80%BC%E6%94%B6%E6%95%9B%E6%88%90%E4%B8%80%E5%8F%A5%E5%8F%AF%E9%AA%8C%E8%AF%81%E7%9A%84%E5%91%BD%E9%A2%98%EF%BC%9A) Claude Skills 架构拆解:渐进披露、运行时与安全沙箱 - Claude 中文 - Claude AI 开发技术社区 [\[5\]](https://cloud.tencent.com/developer/article/2616585#:~:text=Claude%20Skills%E6%98%AFAnthropic%E6%8E%A8%E5%87%BA%E7%9A%84AI%E5%8A%9F%E8%83%BD%E9%9D%A9%E5%91%BD%EF%BC%8C%E5%8F%AF%E5%B0%86%E7%94%A8%E6%88%B7%E4%BD%BF%E7%94%A8AI%E7%9A%84%E4%B9%A0%E6%83%AF%E6%96%87%E4%BB%B6%E5%8C%96%E7%AE%A1%E7%90%86%E3%80%82%E5%AE%83%E8%83%BD%E8%A7%A3%E5%86%B3Claude%E5%81%A5%E5%BF%98%E3%80%81%E9%9C%80%E9%87%8D%E5%A4%8D%E6%8F%90%E7%A4%BA%E8%AF%8D%E7%9A%84%E9%97%AE%E9%A2%98%EF%BC%8C%E5%B0%86%E4%BB%BB%E5%8A%A1%E8%AF%B4%E6%98%8E%E3%80%81%E5%B7%A5%E5%85%B7%E4%BB%A3%E7%A0%81%E7%AD%89%E6%89%93%E5%8C%85%E6%88%90%20) 最近很火爆的Claude Skills到底是个啥?解决什么问题?怎么用! [\[6\]](https://www.facebook.com/iamvista/photos/%E6%9F%90%E5%A4%A9%E6%B7%B1%E5%A4%9C%E6%88%91%E6%AD%A3%E5%9C%A8%E8%B6%95%E4%B8%80%E4%BB%BD%E6%96%87%E4%BB%B6%E5%A4%A9%E5%95%8A%E5%90%8C%E6%A8%A3%E7%9A%84%E6%9E%B6%E6%A7%8B%E5%90%8C%E6%A8%A3%E7%9A%84%E8%AA%9E%E6%B0%A3%E5%90%8C%E6%A8%A3%E7%9A%84%E6%A0%BC%E5%BC%8F%E8%A6%81%E6%B1%82%E4%BD%86%E6%88%91%E5%8F%88%E5%BE%97%E9%87%8D%E6%96%B0%E6%89%93%E4%B8%80%E6%AC%A1%E8%AB%8B%E7%94%A8%E9%80%99%E5%80%8B%E6%A0%BC%E5%BC%8F%E8%AB%8B%E5%85%88%E5%95%8F%E4%B8%89%E5%80%8B%E6%BE%84%E6%B8%85%E5%95%8F%E9%A1%8C%E8%AB%8B%E6%8A%8A%E8%BC%B8%E5%87%BA%E5%88%86%E6%88%90%E5%9B%9B%E6%AE%B5%E8%AB%8B%E9%99%84%E4%B8%8A%E5%8F%AF%E7%9B%B4%E6%8E%A5%E8%B2%BC%E5%88%B0-notion-%E7%9A%84/10162293093624053/#:~:text=,%E6%88%91%E6%80%8E%E9%BA%BC%E5%81%9A%E4%B8%80%E5%80%8B%E5%A5%BDskill%EF%BC%8C%E8%AE%8A%E6%88%90skill) 請先問三個澄清問題、請把輸出分成四段、請附上可直接貼到Notion ... [\[14\]](https://github.com/wensia/xiaohongshu-poster-generator#:~:text=wensia%2Fxiaohongshu,Code%20%2B%20MCP%20%E7%9A%84%E5%B0%8F%E7%BA%A2%E4%B9%A6%E6%98%9F%E5%BA%A7%E5%86%85%E5%AE%B9%E8%87%AA%E5%8A%A8%E5%8C%96%E5%B7%A5%E4%BD%9C%E6%B5%81%EF%BC%8C%E6%94%AF%E6%8C%81%E6%B5%B7%E6%8A%A5%E7%94%9F%E6%88%90%E3%80%81%E7%88%86%E6%96%87%E5%88%86%E6%9E%90%E3%80%81%E5%86%85%E5%AE%B9%E5%8F%91%E5%B8%83%E7%AD%89%E5%8A%9F%E8%83%BD%E3%80%82%20%E5%8A%9F%E8%83%BD%E7%89%B9%E6%80%A7) wensia/xiaohongshu-poster-generator: 小红书星座海报生成器 - GitHub [\[15\]](https://blog.csdn.net/yangshangwei/article/details/156836796#:~:text=%E4%B8%80%E4%B8%AA%E5%8F%AF%E8%90%BD%E5%9C%B0%E7%9A%84%E4%B8%89%E6%AD%A5%E6%B3%95,3%20%E7%AC%AC%E4%B8%89%E6%AD%A5%EF%BC%9A%E6%8A%8AClaude%20%E5%BD%93%E2%80%9C%E6%90%AD%E5%AD%90%E2%80%9D%EF%BC%8C%E8%80%8C%E4%B8%8D%E6%98%AF%E6%90%9C%E7%B4%A2%E6%A1%86) LLM - 从Prompt 到Skills_skills 市场 - CSDN博客 [\[18\]](https://hbwdj.gov.cn/appbdetail-imqqsmrp9897358.d#:~:text=%E9%AA%97%E4%BD%A0%E7%9A%84%EF%BC%8C%E5%85%B6%E5%AE%9EAI%E6%A0%B9%E6%9C%AC%E4%B8%8D%E9%9C%80%E8%A6%81%E9%82%A3%E4%B9%88%E5%A4%9A%E6%8F%90%E7%A4%BA%E8%AF%8D%E3%80%82%20%E4%BD%A0%E5%8F%AA%E9%9C%80%E8%A6%81%E8%B0%83%E7%94%A8AI%20%E6%9C%AC%E8%BA%AB%E7%9A%84%E2%80%9CSkill%20Creator%E2%80%9D%E6%8A%80%E8%83%BD%EF%BC%8C%E7%94%A8%E4%BD%A0%E7%9A%84%E8%AF%AD%E8%A8%80%E6%8F%8F%E8%BF%B0%E8%87%AA%E5%B7%B1%E7%9A%84%E9%9C%80%E6%B1%82%EF%BC%8C%E8%AE%A9AI%E8%87%AA%E5%8A%A8%E5%B8%AE%E4%BD%A0%E7%94%9F%E6%88%90%E4%B8%80%E9%97%A8%E6%8A%80%E8%83%BD%EF%BC%8C%E4%BD%BF%E7%94%A8%E8%B5%B7%E6%9D%A5%E9%9D%9E%E5%B8%B8%E5%8F%8B%E5%A5%BD%EF%BC%8CAI%E4%BC%9A%E4%B8%80%E6%AD%A5%E6%AD%A5%E5%BC%95%E5%AF%BC%E4%BD%A0%E8%AF%B4%E5%87%BA%E4%BD%A0%E7%9A%84%E9%9C%80%E6%B1%82%EF%BC%8C%E4%BD%A0%E5%8F%AA%20) 骗你的,其实AI根本不需要那么多提示词。 [\[19\]](https://lilys.ai/zh/notes/claude-code-20251031/claude-code-skills-automate-everything#:~:text=%E6%83%B3%20%E8%87%AA%E5%8A%A8%E5%8C%96%E4%BD%A0%E6%89%80%E5%81%9A%E7%9A%84%E4%B8%80%E5%88%87%20%E5%90%97%EF%BC%9F%E5%AD%A6%E4%B9%A0%E5%A6%82%E4%BD%95%E5%88%A9%E7%94%A8%20%E5%85%8B%E5%8A%B3%E5%BE%B7%C2%B7%E7%A7%91%E5%BE%B7%E6%8A%80%E8%83%BD%20%E5%88%9B%E5%BB%BA%E8%87%AA%E5%AE%9A%E4%B9%89%E5%B7%A5%E4%BD%9C%E6%B5%81%E3%80%82%E6%9C%AC%E5%86%85%E5%AE%B9%E6%8F%AD%E7%A4%BA%E4%BA%86,%E5%85%AD%E6%AD%A5MASTER%E6%A1%86%E6%9E%B6%EF%BC%8C%E6%95%99%E4%BD%A0%E5%A6%82%E4%BD%95%E9%80%9A%E8%BF%87%20%E8%BF%AD%E4%BB%A3%E5%8F%8D%E9%A6%88%20%E8%AE%AD%E7%BB%83AI%E3%80%82%E6%8E%8C%E6%8F%A1%E6%AD%A4%E6%96%B9%E6%B3%95%EF%BC%8C%E4%BD%A0%E5%B0%B1%E8%83%BD%E5%B0%86%E9%87%8D%E5%A4%8D%E4%BB%BB%E5%8A%A1%E8%BD%AC%E5%8C%96%E4%B8%BA%E9%AB%98%E6%95%88%E7%9A%84%20%E7%B3%BB%E7%BB%9F%E5%8C%96%E6%8A%80%E8%83%BD%EF%BC%8C%E5%AE%9E%E7%8E%B0%E5%B7%A5%E4%BD%9C%E6%95%88%E7%8E%87%E7%9A%84%E6%8C%87%E6%95%B0%E7%BA%A7%E6%8F%90%E5%8D%87%E3%80%82) [\[20\]](https://lilys.ai/zh/notes/claude-code-20251031/claude-code-skills-automate-everything#:~:text=4.%20MASTER%E6%A1%86%E6%9E%B6%EF%BC%9A%E7%AC%AC%E4%BA%8C%E9%98%B6%E6%AE%B5%E2%80%94%E2%80%94%E7%B3%BB%E7%BB%9F%E5%8C%96%E4%B8%BA%E6%8A%80%E8%83%BD%20) [\[21\]](https://lilys.ai/zh/notes/claude-code-20251031/claude-code-skills-automate-everything#:~:text=1.%20%E7%B3%BB%E7%BB%9F%E5%8C%96%E7%9B%AE%E6%A0%87%EF%BC%9A%E5%B0%86%E5%AD%A6%E5%88%B0%E7%9A%84%E6%89%80%E6%9C%89%E5%85%B3%E9%94%AE%E7%BB%8F%E9%AA%8C%E6%95%99%E8%AE%AD%E8%BD%AC%E5%8C%96%E4%B8%BA%20%E7%B3%BB%E7%BB%9F%E5%8C%96%E6%8A%80%E8%83%BD%E3%80%82,48) [\[22\]](https://lilys.ai/zh/notes/claude-code-20251031/claude-code-skills-automate-everything#:~:text=match%20at%20L3327%205,57) 克劳德代码技能:自动化你的所有操作 [\[31\]](https://lilys.ai/zh/notes/claude-skills-20251022/no-code-ai-workflow-claude-skills#:~:text=Skill%20%E5%9F%BA%E7%A1%80%E7%BB%93%E6%9E%84%EF%BC%9A%E6%9C%80%E7%AE%80%E5%8D%95%E6%83%85%E5%86%B5%E4%B8%8B%EF%BC%8CSkill%20%E6%98%AF%E4%B8%80%E4%B8%AA%E5%8C%85%E5%90%ABSkill%20Markdown%20%E6%96%87%E4%BB%B6%E7%9A%84%E7%9B%AE%E5%BD%95,36%5D) 无需编写代码也能定制AI 工作流?Claude Skills 让你的AI 更懂你 [\[32\]](https://github.com/0xfnzero/AI-Code-Tutorials#:~:text=%E4%BB%8E%E9%9B%B6%E5%9F%BA%E7%A1%80%E5%88%B0%E9%AB%98%E7%BA%A7%E5%BA%94%E7%94%A8%EF%BC%8C%E7%B3%BB%E7%BB%9F%E5%AD%A6%E4%B9%A0Claude%20Code%EF%BC%8C%E6%8E%8C%E6%8F%A1AI%20%E8%BE%85%E5%8A%A9%E7%BC%96%E7%A8%8B%E6%8A%80%E8%83%BD%EF%BC%8C%E6%8F%90%E5%8D%87%E5%BC%80%E5%8F%91%E6%95%88%E7%8E%8710%20%E5%80%8D%20,md%E3%80%81%E5%B7%A5%E5%85%B7%E6%9D%83%E9%99%90%E3%80%81gh%20CLI%EF%BC%89%3B%20%E7%BB%99Claude%20%E6%9B%B4%E5%A4%9A%E5%B7%A5%E5%85%B7%EF%BC%88bash%E3%80%81MCP) 0xfnzero/AI-Code-Tutorials: 从零基础到高级应用,系统学习Claude ... [\[33\]](https://news.qq.com/rain/a/20260107A02N2N00#:~:text=Skills%20%E5%B0%B1%E6%98%AF%E7%BB%99Claude%20%E7%9A%84) 给AI装个技能包:Skills是什么-腾讯新闻 - QQ [\[34\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=match%20at%20L93%20,side%20rendering%20and) [\[35\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=match%20at%20L96%20,No%20magic%2C%20no) [\[36\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=,No%20magic%2C%20no) [\[37\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=,No%20magic%2C%20no) [\[38\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=Claude%20can%20describe%20architecture%20well%2C,com%20which%20I%20then) [\[39\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=match%20at%20L316%20Give%20Claude,context%2C%20the%20better%20the%20suggestions) [\[40\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=match%20at%20L322%20When%20Claude,things%20you%E2%80%99ve%20already%20decided%20against) [\[42\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=After%20long%20chats%2C%20Claude%20noticeably,you%20have%20done%20so%20far) [\[45\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=if%20err%20%3A%3D%20godotenv,) [\[46\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=match%20at%20L236%20This%20is,JavaScript%20pretending%20to%20be%20HTML) [\[47\]](https://medium.com/comsystoreply/learning-go-and-vue-with-claude-ai-as-my-pair-programmer-b2d634e291eb#:~:text=3,getting%20lost%20in%20tutorial%20hell) Learning Go and Vue with Claude AI as my Pair Programmer | comsystoreply [\[41\]](https://www.cnblogs.com/treasury-manager/p/19217990#:~:text=OpenCode%20%E4%B8%80%E4%B8%AA%E7%A5%9E%E5%A5%87%E7%9A%84CLI%20%EF%BC%88%E5%8F%AF%E4%BB%A5%E8%9E%8D%E5%90%88Claude%20Code%2C%20iFLow,code%29%20%E2%86%90%20%E8%87%AA%E5%8A%A8%E4%BD%BF%E7%94%A8oracle%20%E7%9A%84%E6%A8%A1%E5%9E%8B%E2%86%93%20%E8%B0%83%E7%94%A8%40explore) [\[80\]](https://www.cnblogs.com/treasury-manager/p/19217990#:~:text=OpenCode%20%E4%B8%80%E4%B8%AA%E7%A5%9E%E5%A5%87%E7%9A%84CLI%20%EF%BC%88%E5%8F%AF%E4%BB%A5%E8%9E%8D%E5%90%88Claude%20Code%2C%20iFLow,code%29%20%E2%86%90%20%E8%87%AA%E5%8A%A8%E4%BD%BF%E7%94%A8oracle%20%E7%9A%84%E6%A8%A1%E5%9E%8B%E2%86%93%20%E8%B0%83%E7%94%A8%40explore) OpenCode 一个神奇的CLI (可以融合Claude Code, iFLow ... - 博客园 [\[43\]](https://opencode.ai/docs#:~:text=%2Finit) [\[44\]](https://opencode.ai/docs#:~:text=You%20can%20ask%20OpenCode%20to,explain%20the%20codebase%20to%20you) [\[52\]](https://opencode.ai/docs#:~:text=1) [\[53\]](https://opencode.ai/docs#:~:text=OpenCode%20is%20an%20open%20source,desktop%20app%2C%20or%20IDE%20extension) [\[58\]](https://opencode.ai/docs#:~:text=Configure) [\[59\]](https://opencode.ai/docs#:~:text=If%20you%20are%20new%20to,verified%20by%20the%20OpenCode%20team) [\[60\]](https://opencode.ai/docs#:~:text=Next%2C%20initialize%20OpenCode%20for%20the,by%20running%20the%20following%20command) [\[61\]](https://opencode.ai/docs#:~:text=This%20will%20get%20OpenCode%20to,file%20in%20the%20project%20root) [\[62\]](https://opencode.ai/docs#:~:text=Make%20changes) [\[63\]](https://opencode.ai/docs#:~:text=1) [\[64\]](https://opencode.ai/docs#:~:text=Now%20let%E2%80%99s%20describe%20what%20we,want%20it%20to%20do) [\[66\]](https://opencode.ai/docs#:~:text=2) [\[67\]](https://opencode.ai/docs#:~:text=2) [\[68\]](https://opencode.ai/docs#:~:text=3) [\[69\]](https://opencode.ai/docs#:~:text=Once%20you%20feel%20comfortable%20with,hitting%20the%20Tab%20key%20again) [\[70\]](https://opencode.ai/docs#:~:text=Tip) [\[71\]](https://opencode.ai/docs#:~:text=Let%E2%80%99s%20say%20you%20ask%20OpenCode,to%20make%20some%20changes) [\[72\]](https://opencode.ai/docs#:~:text=Share) [\[73\]](https://opencode.ai/docs#:~:text=You%20want%20to%20give%20OpenCode,junior%20developer%20on%20your%20team) [\[74\]](https://opencode.ai/docs#:~:text=We%20need%20to%20add%20authentication,look%20at%20how%20this%20is) [\[75\]](https://opencode.ai/docs#:~:text=without%20having%20to%20review%20the,plan%20first) Intro | OpenCode [\[48\]](https://zhuanlan.zhihu.com/p/1991170184573122515#:~:text=OpenCode%20%E6%98%AF%E4%B8%80%E4%B8%AAAI%20%E7%BC%96%E7%A8%8B%E5%8A%A9%E6%89%8B%EF%BC%8C%E8%B7%91%E5%9C%A8%E4%BD%A0%E7%9A%84%E7%BB%88%E7%AB%AF%E9%87%8C%EF%BC%88%E5%B0%B1%E6%98%AF%E9%82%A3%E4%B8%AA%E9%BB%91%E8%89%B2%E7%AA%97%E5%8F%A3%EF%BC%89%E3%80%82%20%E4%BD%A0%E8%B7%9F%E5%AE%83%E8%AF%B4%E8%AF%9D%EF%BC%8C%E5%AE%83%E5%B0%B1%E5%B8%AE%E4%BD%A0%E5%86%99%E4%BB%A3%E7%A0%81%E3%80%82%20,%E2%86%92%20%E5%AE%83%E6%94%B9) OpenCode + Oh My OpenCode 一份老奶奶都能看懂的AI 编程指南 [\[49\]](https://github.com/anomalyco/opencode#:~:text=close%20and%20pricing%20will%20drop,what%27s%20possible%20in%20the%20terminal) [\[51\]](https://github.com/anomalyco/opencode#:~:text=OpenCode%20includes%20two%20built,key) [\[54\]](https://github.com/anomalyco/opencode#:~:text=%2A%20100,what%27s%20possible%20in%20the%20terminal) [\[55\]](https://github.com/anomalyco/opencode#:~:text=%2A%20100,push%20the%20limits%20of%20what%27s) [\[56\]](https://github.com/anomalyco/opencode#:~:text=Installation) [\[57\]](https://github.com/anomalyco/opencode#:~:text=npm%20i%20,Any%20OS) [\[65\]](https://github.com/anomalyco/opencode#:~:text=%2A%20build%20,unfamiliar%20codebases%20or%20planning%20changes) GitHub - anomalyco/opencode: The open source coding agent. [\[50\]](https://cloud.tencent.com/developer/article/2574516#:~:text=Claude%20Code%E4%BB%A3%E7%A0%81%E8%A7%84%E8%8C%83%E5%AE%88%E6%8A%A4%E8%80%85%E5%AD%90%E4%BB%A3%E7%90%86%E5%AE%9E%E6%88%98%E6%8C%87%E5%8D%97%20,Code%E4%BB%A3%E7%A0%81%E8%A7%84%E8%8C%83%E5%AD%90%E4%BB%A3%E7%90%86%E6%98%AFAI%E9%A9%B1%E5%8A%A8%E7%9A%84%E4%BB%A3%E7%A0%81%E8%B4%A8%E9%87%8F%E7%AE%A1%E5%AE%B6%EF%BC%8C%E8%83%BD%E8%87%AA%E5%8A%A8%E7%BB%9F%E4%B8%80%E5%9B%A2%E9%98%9F%E4%BB%A3%E7%A0%81%E9%A3%8E%E6%A0%BC%E3%80%81%E6%89%A7%E8%A1%8C%E5%91%BD%E5%90%8D%E8%A7%84%E8%8C%83%E3%80%81%E6%A3%80%E6%9F%A5%E6%B5%8B%E8%AF%95%E8%A6%86%E7%9B%96%E7%8E%87%E3%80%82%E9%80%9A%E8%BF%87ESLint%E3%80%81Prettier%E7%AD%89%E5%B7%A5%E5%85%B7%E9%85%8D%E7%BD%AE%EF%BC%8C) Claude Code代码规范守护者子代理实战指南 - 腾讯云 [\[76\]](https://blog.csdn.net/u012094427/article/details/148866474#:~:text=%E4%BB%8A%E5%A4%A9%E6%88%91%E4%BB%AC%E8%A6%81%E8%81%8A%E7%9A%84%E7%A1%AC%E6%A0%B8%E8%AF%9D%E9%A2%98%EF%BC%8C%E6%98%AF%E4%B8%AA%E8%AE%A9%E6%9E%81%E5%AE%A2%E4%BB%AC%E9%A2%A4%E6%8A%96%E3%80%81%E8%AE%A9%E7%A8%8B%E5%BA%8F%E5%91%98%E4%BB%AC%E5%B0%96%E5%8F%AB%EF%BC%8C%E8%AE%A9%E5%86%99%E4%BB%A3%E7%A0%81%E7%88%BD%E5%88%B0%E9%A3%9E%E8%B5%B7%E7%9A%84%E5%AD%98%E5%9C%A8%E2%80%94%E2%80%94OpenCode%EF%BC%8C%E5%BC%80%E6%BA%90AI%E7%BB%88%E7%AB%AF%E7%BC%96%E7%A0%81%E5%8A%A9%E6%89%8B%E3%80%82%20%E5%9C%A8AI%E5%85%A8%E8%83%BD%E5%86%99%E4%BB%A3%E7%A0%81%E3%80%81Copilot%E5%AE%B6%E5%A4%A7%E4%B8%9A%E5%A4%A7%E3%80%81%20) 【爆款长文】终端里的AI编程老司机--全面解读OpenCode! 原创 [\[77\]](https://x.com/Nateemerson/status/2002043382953288046/photo/1#:~:text=YeonGyu,and%20practices%20for%20agentic%20coding) YeonGyu-Kim built Oh My Opencode (OmO), an OC plugin ... [\[78\]](https://github.com/code-yeongyu/oh-my-opencode#:~:text=The%20Best%20Agent%20Harness,Agent%20that%20codes%20like%20you) GitHub - code-yeongyu/oh-my-opencode: The Best Agent Harness. Meet Sisyphus: The Batteries-Included Agent that codes like you. [\[79\]](https://post.smzdm.com/p/aog8xq7m#:~:text=%E4%B8%80%E4%BA%BA%E6%8A%B5%E4%B8%80%E4%B8%AA%E5%BC%80%E5%8F%91%E5%9B%A2%E9%98%9F%E7%9A%84AI%E7%BC%96%E7%A8%8B%E7%A5%9E%E5%99%A8%E5%AE%8C%E5%85%A8%E6%8C%87%E5%8D%97%20,%E5%9C%A8%E6%88%90%E6%9C%AC%E6%8E%A7%E5%88%B6%E6%96%B9%E9%9D%A2%EF%BC%8C%E8%AF%A5%E7%B3%BB%E7%BB%9F%E6%94%AF%E6%8C%81%E5%A4%9A%E6%A8%A1%E5%9E%8B%E6%B7%B7%E5%90%88%E4%BD%BF%E7%94%A8%E7%AD%96%E7%95%A5%E3%80%82%E9%80%9A%E8%BF%87%E9%85%8D%E7%BD%AE%EF%BC%8C%E5%8F%AF%E4%BB%A5%E5%B0%86) [\[85\]](https://post.smzdm.com/p/aog8xq7m#:~:text=%E5%9C%A8%E6%88%90%E6%9C%AC%E6%8E%A7%E5%88%B6%E6%96%B9%E9%9D%A2%EF%BC%8C%E8%AF%A5%E7%B3%BB%E7%BB%9F%E6%94%AF%E6%8C%81%E5%A4%9A%E6%A8%A1%E5%9E%8B%E6%B7%B7%E5%90%88%E4%BD%BF%E7%94%A8%E7%AD%96%E7%95%A5%E3%80%82%E9%80%9A%E8%BF%87%E9%85%8D%E7%BD%AE%EF%BC%8C%E5%8F%AF%E4%BB%A5%E5%B0%86%20) 一人抵一个开发团队的AI编程神器完全指南 - 什么值得买 [\[81\]](https://github.com/code-yeongyu/oh-my-opencode/blob/dev/README.zh-cn.md#:~:text=oh,%E8%80%81%E5%AE%9E%E8%AF%B4%EF%BC%8C%E7%94%9A%E8%87%B3%E4%B8%8D%E7%94%A8%E8%B4%B9%E5%BF%83%E8%AF%BB%E6%96%87%E6%A1%A3%E3%80%82%E5%8F%AA%E9%9C%80%E5%86%99%E4%BD%A0%E7%9A%84%E6%8F%90%E7%A4%BA%E3%80%82%E5%8C%85%E5%90%AB%27ultrawork%27%20%E5%85%B3%E9%94%AE%E8%AF%8D%E3%80%82Sisyphus%20%E4%BC%9A%E5%88%86%E6%9E%90%E7%BB%93%E6%9E%84%EF%BC%8C%E6%94%B6%E9%9B%86%E4%B8%8A%E4%B8%8B%E6%96%87%EF%BC%8C%E6%8C%96%E6%8E%98%E5%A4%96%E9%83%A8%E6%BA%90%E4%BB%A3%E7%A0%81%EF%BC%8C%E7%84%B6%E5%90%8E%E6%8C%81%E7%BB%AD%E6%8E%A8%E8%BF%9B) oh-my-opencode/README.zh-cn.md at dev - GitHub [\[82\]](https://www.youtube.com/watch?v=twFjLiy2Pmc#:~:text=%E8%A7%86%E9%A2%91%E7%AE%80%E4%BB%8B%EF%BC%9A%20%E6%9C%AC%E6%9C%9F%E8%A7%86%E9%A2%91%E8%AF%A6%E7%BB%86%E6%BC%94%E7%A4%BA%E4%BA%86%E5%A6%82%E4%BD%95%E5%9C%A8Opencode%E4%B8%AD%E4%BD%BF%E7%94%A8%E6%9C%80%E5%BC%BA%E5%BC%80%E6%BA%90%E6%8F%92%E4%BB%B6Oh%20My%20Opencode%EF%BC%88OMO%EF%BC%89%EF%BC%8C%E5%B0%86%E5%8D%95%E4%B8%80AI%E7%BC%96%E7%A8%8B%E5%8A%A9%E6%89%8B%E5%8D%87%E7%BA%A7%E4%B8%BA%E5%A4%9A%E6%A8%A1%E5%9E%8B%E5%8D%8F%E4%BD%9C%E7%9A%84AI%E5%BC%80%E5%8F%91%E5%9B%A2%E9%98%9F%EF%BC%81) Sisyphus智能体指挥多AI协作,复杂项目开发效率倍增!全程零干预 ... [\[83\]](https://x.com/AISuperDomain/status/2009823408301994209#:~:text=OpenCode%E7%9A%84%E6%9C%80%E5%BC%BA%E5%BC%80%E6%BA%90%E6%8F%92%E4%BB%B6oh%20my%20opencode%E7%A1%AE%E5%AE%9E%E9%9D%9E%E5%B8%B8%E5%A5%BD%E7%94%A8%20%E8%A7%86%E9%A2%91%E7%AE%80%E4%BB%8B%EF%BC%9A%20%E6%9C%AC%E6%9C%9F%E8%A7%86%E9%A2%91%E8%AF%A6%E7%BB%86%E6%BC%94%E7%A4%BA%E4%BA%86%E5%A6%82%E4%BD%95%E5%9C%A8Opencode%E4%B8%AD%E4%BD%BF%E7%94%A8%E6%9C%80%E5%BC%BA%E5%BC%80%E6%BA%90%E6%8F%92%E4%BB%B6Oh,My%20Opencode%EF%BC%88OMO%EF%BC%89%EF%BC%8C%E5%B0%86%E5%8D%95%E4%B8%80AI%E7%BC%96%E7%A8%8B%E5%8A%A9%E6%89%8B%E5%8D%87%E7%BA%A7%E4%B8%BA%E5%A4%9A%E6%A8%A1%E5%9E%8B%E5%8D%8F%E4%BD%9C%E7%9A%84AI%E5%BC%80%E5%8F%91%E5%9B%A2%E9%98%9F%EF%BC%81%20%E6%A0%B8%E5%BF%83%E5%86%85%E5%AE%B9%EF%BC%9AOMO%E6%8F%92%E4%BB%B6) OpenCode的最强开源插件oh my opencode确实非常好用 [\[84\]](https://blog.csdn.net/gitblog_00895/article/details/144862506#:~:text=oh,) oh-my-opencode高级用法:自定义钩子和插件系统详解 - CSDN博客