SuperAPI AI Directory

my-init

General & other

View repository

Updated 10-16 09:30

About this skill

Init - Initialize project AI context

Scan the project structure and generate a hierarchical CLAUDE.md document index.

Usage

/my-init [项目简述] [--modules]

Parameter description:

  • 项目简述: optional, a brief description of the project
  • --modules: optional, when this parameter is added, generate module-level CLAUDE.md files

Execution workflow

Step 1: Get timestamp + scan project

Execute in parallel:

// 1. 获取时间戳
Bash({ command: "powershell -Command \"Get-Date -Format 'yyyy-MM-dd HH:mm'\"" })
// Unix: date "+%Y-%m-%d %H:%M"

// 2. 扫描项目结构(只读,返回结构信息)
Task({
  subagent_type: "Explore",
  prompt: `分析项目结构并返回 JSON 格式报告:
{
  "name": "项目名称(从 package.json/pyproject.toml 或目录名推断)",
  "tech": { "language": "", "framework": "", "build": "" },
  "entry": ["入口文件路径"],
  "modules": [
    { "path": "模块路径", "name": "模块名", "role": "职责", "files": ["关键文件"] }
  ],
  "commands": { "install": "", "dev": "", "test": "" }
}
跳过:node_modules、dist、build、.git、__pycache__`,
  description: "扫描项目结构"
})

Step 2: Generate root-level CLAUDE.md

Use the Write toolCreate in the project root directory CLAUDE.md:

# {项目名称}

> 生成时间:{TIMESTAMP}

## 概述

{根据扫描结果写 1-2 句项目简述}

## 技术栈

- 语言:{language}
- 框架:{framework}
- 构建:{build}

## 项目结构

```mermaid
graph TD
    ROOT[{项目名}]
    ROOT --> M1[{模块1}]
    ROOT --> M2[{模块2}]

Module index

Module Path Responsibility
{name} {path} {role}

Entry file

  • {entry} - {description}

Quick commands

{commands.install}
{commands.dev}
{commands.test}

### 步骤 3:生成模块级 CLAUDE.md(仅当指定 `--modules` 参数时)

**跳过条件**:如果用户未指定 `--modules` 参数,跳过此步骤。

**对每个模块**,用 Write 工具在其目录下创建 `CLAUDE.md`:

```markdown
# {模块名}

> [← 返回根目录](../CLAUDE.md) | 更新:{TIMESTAMP}

## 职责

{role}

## 关键文件

| 文件 | 职责 |
|------|------|
| `{file}` | {描述} |

## 依赖关系

- 依赖:{分析 import 语句}
- 被依赖:{分析其他模块的 import}

Step 4: Output summary

## 初始化完成

### 生成文档
- ✅ `./CLAUDE.md`
- ✅ 模块级:{N} 个(仅当使用 `--modules` 参数时显示)

### 识别模块
1. `{path}` - {role}

### 推荐下一步
- [ ] 检查文档准确性
- [ ] 补充业务逻辑
- [ ] 如需模块级文档,使用 `/my-init --modules` 重新生成

Rules

  1. Explore is read-only - subagents only return analysis results, do not write files
  2. Main process writes files - you (the Claude executing the skill) generate documents with the Write tool
  3. Only write documentation - do not modify source code
  4. Skip generated artifacts - node_modules、dist、build、__pycache__、.git
  5. Module identification - directories containing __init__.py(Python) or index.*(JS/TS)

Mermaid conventions

  • Root level:graph TD(top-down)
  • Module level:graph LR(left-to-right)
  • Node IDs use camelCase naming