Files
test-repo/CLAUDE.md
T
sanbuphy a4b583b13f feat(docs): integrate version2 curriculum and stage-3 updates
概要
- 将 version2 分支的课程结构重构、第三阶段章节新增、示例资源迁移、高级 RAG 文档与 Vercel 部署配置等整合为 main 上的一次汇总提交

内容导航与 README 调整
- 更新 README 的总体介绍文案,引入“第零阶段 + 第一到第三阶段”的完整学习路径描述
- 将原先的“三阶段实战路径”说明替换为新版分阶段描述,突出从小游戏到跨平台复杂应用的学习节奏
- 删除已过时的“第二次更新将在分支 version2 合并到主分支”的提示,改为直接以 main 为主线
- 统一 README 顶部标题和排版风格,保证中英文导航、徽章展示等视觉结构一致

课程结构与章节导航更新
- 调整 docs 目录下的学习阶段导航结构,使 README 中的导航表与各 stage 实际目录对齐
- 补全并创建 stage-3 相关章节入口文件,用于承载高级阶段的课程内容
- 新增或更新以下章节入口:
  - 高级核心技能:
    - docs/stage-3/core-skills/3.1-mcp-claudecode-skills/index.md
    - docs/stage-3/core-skills/3.2-long-running-tasks/index.md
  - 多平台开发:
    - docs/stage-3/cross-platform/3.3-wechat-miniprogram/index.md
    - docs/stage-3/cross-platform/3.4-wechat-miniprogram-backend/index.md
    - docs/stage-3/cross-platform/3.5-android-app/index.md
    - docs/stage-3/cross-platform/3.6-ios-app/index.md
  - 个人品牌:
    - docs/stage-3/personal-brand/3.7-personal-website-blog/index.md
- 保持 stage-0、stage-1、stage-2 既有章节结构不变的前提下,对导航表格进行排版和链接校正,使整体课程地图清晰、可点击

示例与图片资源重组
- 将原先位于 docs/examples/example1/images/ 下的微信小程序示例图片,整体迁移到 stage-3 的正式课程路径中:
  - 目标路径:docs/stage-3/3.3-how-to-build-a-wechat-miniprogram/example1/images/
- 通过 rename 方式保留 git 历史关系,避免图片资源被视为完全新增,从而方便后续追踪
- 为微信小程序示例新增 index 页面:
  - docs/stage-3/3.3-how-to-build-a-wechat-miniprogram/example1/index.md
- 使该示例在“高级三:多平台开发:如何构建微信小程序”章节中有清晰的入口,对应实际实战内容

高级 RAG 与 AI 进阶文档
- 新增一篇系统介绍 RAG 的高级文档:
  - docs/stage-3/ai-advanced/3.a1-rag-introduction/extra5-what-is-rag-and-how-does-it-work-and-future.md
- 覆盖内容包括:RAG 的基本概念、典型架构、工作流程以及未来演进方向,为第三阶段的复杂应用提供知识检索基础
- 配套引入多张插图,帮助读者从架构图和流程视角理解 RAG:
  - docs/stage-3/ai-advanced/3.a1-rag-introduction/images/image1.png ~ image15.png

部署与工程配置
- 新增 vercel.json 配置文件,为项目在 Vercel 上的部署提供基础配置
  - 明确文档构建产物的输出路径和静态站点托管方式
  - 为之后的一键部署和自动化预览打下基础

依赖与锁文件更新
- 调整 package.json 中与新版文档结构和部署相关的配置,保持脚本和依赖与当前课程形态同步
- 更新 package-lock.json,以反映最新的依赖树和版本锁定状态
- 保证在执行 npm install / npm run build 时,依赖环境与 version2 中的实际使用情况一致

兼容性与行为说明
- 该提交通过 npm run build 验证,确保在整合 version2 内容后,VitePress 构建过程正常完成
- main 分支上的历史被压缩为一条有语义的“第二次大更新”提交,详细的开发过程仍保留在 version2 分支,用于后续需要时回溯
2026-01-12 12:25:44 +08:00

5.8 KiB

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

Project Overview

This is Easy-Vibe, an educational curriculum for learning AI Vibe Coding from zero to advanced levels. It's a documentation-based project using Docsify to serve educational content about AI-assisted software development.

The curriculum follows a progressive four-stage structure:

  • Stage 0 (幼儿园): Introduction to AI programming through games
  • Stage 1 (AI 产品经理): Building AI-powered web application prototypes
  • Stage 2 (初中级开发工程师): Full-stack development with databases and deployment
  • Stage 3 (高级开发工程师): Cross-platform development (WeChat mini-programs, Android apps, MCP)

Development Commands

Start Local Documentation Server

npm install      # Install dependencies (first time only)
npm run dev      # Start docsify server on port 3000
# or
npm start        # Same as npm run dev

The documentation will be available at http://localhost:3000

Build/Run Commands

  • npm run dev - Serves the documentation using docsify-cli on port 3000
  • No build step required - docsify serves markdown files directly

Project Architecture

Directory Structure

easy-vibe/
├── docs/                    # Main documentation content (served by docsify)
│   ├── index.html          # Docsify configuration and plugins
│   ├── _sidebar.md         # Sidebar navigation structure
│   ├── README.md           # Homepage (symlink to root README.md)
│   ├── stage-0/            # Stage 0 content (幼儿园)
│   ├── stage-1/            # Stage 1 content (AI 产品经理)
│   ├── stage-2/            # Stage 2 content (初中级开发工程师)
│   ├── stage-3/            # Stage 3 content (高级开发工程师)
│   ├── appendix/           # Reference materials (AI capability dictionary)
│   ├── examples/           # Practical examples and tutorials
│   ├── extra/              # Additional knowledge (Git, API, RAG, etc.)
│   └── project/            # Legacy project documentation
├── assets/                 # Images and static assets (symlinked in docs/)
├── package.json           # Project dependencies and scripts
└── README.md              # Project overview and contribution guide

Content Organization

Each stage follows a numbered chapter structure:

stage-{N}/
└── {N}.{M}-{chapter-name}/
    ├── index.md          # Main content file
    └── images/           # Chapter-specific images

Example: stage-1/1.1-introduction-to-ai-ide/index.md

Documentation System (Docsify)

The project uses Docsify with the following key configuration in docs/index.html:

  • Single Sidebar: loadSidebar: true with /_sidebar.md alias for all routes
  • Search: Full-text search across all documentation
  • Dark Mode: Theme toggle with localStorage persistence
  • Image Viewer: Viewer.js for image zoom/rotate/flip
  • Code Copy: Copy code button on all code blocks
  • Pagination: Previous/Next navigation between chapters
  • Word Count: Chinese word count display on each page

Sidebar Management

The sidebar is defined in docs/_sidebar.md. When adding new chapters:

  1. Add the chapter entry to the appropriate stage section
  2. Follow the existing pattern: * [Chapter Title](stage-{N}/{chapter-dir}/index.md)
  3. Ensure relative paths match the actual directory structure

Asset Management

  • Static assets are in /assets/ at the root level
  • Chapter-specific images go in stage-{N}/{chapter-dir}/images/
  • Images are referenced with relative paths from the markdown file location
  • Global CSS limits image display to max 80% width, 300px height with centered alignment

Legacy Content Structure

The project maintains three legacy sections in the sidebar for backward compatibility:

  1. Project 文档 (project/): Older chapter-based tutorials
  2. Extra 扩展知识 (extra/): Supplementary topics (Git, APIs, RAG, deployment)
  3. Examples 实战案例 (examples/): Practical tutorials (some moved to stage-3)

When updating content, prefer integrating into the stage structure over adding to legacy sections.

Content Guidelines

Writing New Chapters

  1. Create directory: docs/stage-{N}/{N}.{M}-{chapter-name}/
  2. Create index.md with chapter content
  3. Add images subdirectory if needed: images/
  4. Update docs/_sidebar.md with the new entry
  5. Follow Chinese language conventions (this is a Chinese curriculum)

Content Status Markers

In README.md, use these status indicators:

  • Completed
  • 🚧 In progress/Under construction

File Naming Conventions

  • Use kebab-case for directories: 1.1-introduction-to-ai-ide
  • Use index.md for primary content files
  • Images use descriptive names in their respective chapter directories

Permissions

The project has configured bash permissions in .claude/settings.local.json:

  • File operations: which, find, mv, tree, cat, curl, lsof
  • Process management: xargs ps, kill
  • Development: npm run dev

Key Context for Development

  • Educational Focus: This is curriculum content, not application code
  • Target Audience: Beginners to advanced developers learning AI-assisted programming
  • Language: Primary content is in Chinese
  • No Build Pipeline: Docsify serves markdown directly, no compilation needed
  • Git Workflow: Content changes should preserve formatting and structure
  • Asset Paths: Always use relative paths from markdown file location

When making changes:

  • Preserve the Docsify configuration in docs/index.html
  • Maintain sidebar structure consistency
  • Test locally with npm run dev before committing
  • Check that image links work correctly
  • Ensure dark mode styles are not broken