agent-folder-init

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Agent Folder Init

Agent 文件夹初始化

Create a comprehensive

.agents/

folder structure for AI-first development workflows.

为AI优先的开发工作流创建一个全面的

.agents/

文件夹结构。

Purpose

用途

This skill scaffolds a complete AI agent documentation system including:

Session tracking (daily files)
Task management
Coding standards and rules
Architecture decision records
Security checklists
SOPs for common workflows
Agent config folders (.claude, .codex, .cursor) with commands, rules, and agents

此技能可搭建完整的AI Agent文档系统，包括：

会话跟踪（每日文件）
任务管理
编码标准与规则
架构决策记录
安全检查表
常见工作流的SOP
Agent配置文件夹（.claude、.codex、.cursor），包含命令、规则和Agent

When to Use

使用时机

Use this skill when:

Starting a new project that will use AI coding assistants
Setting up AI-first development workflows
Migrating an existing project to use structured AI documentation

在以下场景使用此技能：

启动将使用AI编码助手的新项目时
搭建AI优先的开发工作流时
将现有项目迁移至结构化AI文档时

Usage

使用方法

Run the scaffold script:

bash

python3 ~/.claude/skills/agent-folder-init/scripts/scaffold.py --help

运行脚手架脚本：

bash

python3 ~/.claude/skills/agent-folder-init/scripts/scaffold.py --help

Basic usage

python3 ~/.claude/skills/agent-folder-init/scripts/scaffold.py
--root /path/to/project
--name "My Project"

With custom options

python3 ~/.claude/skills/agent-folder-init/scripts/scaffold.py
--root /path/to/project
--name "My Project"
--tech "nextjs,nestjs"
--allow-outside

undefined

python3 ~/.claude/skills/agent-folder-init/scripts/scaffold.py
--root /path/to/project
--name "My Project"
--tech "nextjs,nestjs"
--allow-outside

undefined

Generated Structure

生成的结构

Documentation (.agents/)

文档（.agents/）

.agents/
├── README.md                    # Navigation hub
├── SYSTEM/
│   ├── README.md
│   ├── RULES.md                 # Coding standards
│   ├── ARCHITECTURE.md          # What's implemented
│   ├── SUMMARY.md               # Current state
│   ├── PRD.md                   # Product Requirements Document
│   ├── ENTITIES.md              # Entity documentation
│   ├── ai/
│   │   ├── SESSION-QUICK-START.md
│   │   ├── SESSION-DOCUMENTATION-PROTOCOL.md
│   │   └── USER-PREFERENCES.md
│   ├── architecture/
│   │   ├── DECISIONS.md         # ADRs
│   │   └── PROJECT-MAP.md
│   ├── critical/
│   │   ├── CRITICAL-NEVER-DO.md
│   │   └── CROSS-PROJECT-RULES.md
│   └── quality/
│       └── SECURITY-CHECKLIST.md
├── TASKS/
│   ├── README.md
│   └── INBOX.md
├── SESSIONS/
│   ├── README.md
│   └── TEMPLATE.md
├── SOP/
│   └── README.md
├── EXAMPLES/
│   └── README.md
└── FEEDBACK/
    └── README.md

.agents/
├── README.md                    # Navigation hub
├── SYSTEM/
│   ├── README.md
│   ├── RULES.md                 # Coding standards
│   ├── ARCHITECTURE.md          # What's implemented
│   ├── SUMMARY.md               # Current state
│   ├── PRD.md                   # Product Requirements Document
│   ├── ENTITIES.md              # Entity documentation
│   ├── ai/
│   │   ├── SESSION-QUICK-START.md
│   │   ├── SESSION-DOCUMENTATION-PROTOCOL.md
│   │   └── USER-PREFERENCES.md
│   ├── architecture/
│   │   ├── DECISIONS.md         # ADRs
│   │   └── PROJECT-MAP.md
│   ├── critical/
│   │   ├── CRITICAL-NEVER-DO.md
│   │   └── CROSS-PROJECT-RULES.md
│   └── quality/
│       └── SECURITY-CHECKLIST.md
├── TASKS/
│   ├── README.md
│   └── INBOX.md
├── SESSIONS/
│   ├── README.md
│   └── TEMPLATE.md
├── SOP/
│   └── README.md
├── EXAMPLES/
│   └── README.md
└── FEEDBACK/
    └── README.md

Agent Configs

Agent 配置

.claude/
├── commands/                    # Slash commands (project-specific)
│   ├── start.md
│   ├── end.md
│   ├── new-session.md
│   ├── commit-summary.md
│   ├── code-review.md
│   ├── bug.md
│   ├── quick-fix.md
│   ├── refactor-code.md
│   ├── inbox.md
│   ├── task.md
│   ├── validate.md
│   └── clean.md
├── agents/                      # Specialized agents (project-specific)
│   ├── senior-backend-engineer.md
│   └── senior-frontend-engineer.md
└── skills/                      # Project-specific skills

.codex/
├── commands/
└── skills/

.cursor/
└── commands/

Note: Agent configs (agents/, commands/) are copied from the library root (e.g.,

~/.claude/agents/

) to ensure projects get the latest version. Rules are NOT copied because they're inherited from

~/.claude/rules/

at the library/personal level - this prevents duplication and ensures all projects use consistent rules.

.claude/
├── commands/                    # Slash commands (project-specific)
│   ├── start.md
│   ├── end.md
│   ├── new-session.md
│   ├── commit-summary.md
│   ├── code-review.md
│   ├── bug.md
│   ├── quick-fix.md
│   ├── refactor-code.md
│   ├── inbox.md
│   ├── task.md
│   ├── validate.md
│   └── clean.md
├── agents/                      # Specialized agents (project-specific)
│   ├── senior-backend-engineer.md
│   └── senior-frontend-engineer.md
└── skills/                      # Project-specific skills

.codex/
├── commands/
└── skills/

.cursor/
└── commands/

注意：Agent配置（agents/、commands/）会从库根目录（例如

~/.claude/agents/

）复制，以确保项目获取最新版本。规则不会被复制，因为它们会从库/个人级别的

~/.claude/rules/

继承——这避免了重复，并确保所有项目使用一致的规则。

Root Files

根目录文件

```
AGENTS.md
```
- Points to
```
.agents/README.md
```
```
CLAUDE.md
```
- Claude-specific entry point
```
CODEX.md
```
- Codex-specific entry point
```
.editorconfig
```
- Editor configuration

```
AGENTS.md
```
- 指向
```
.agents/README.md
```
```
CLAUDE.md
```
- Claude专属入口
```
CODEX.md
```
- Codex专属入口
```
.editorconfig
```
- 编辑器配置

Key Patterns

核心模式

Naming Conventions

命名规范

Top-level directories: ALL-CAPS (
```
SYSTEM/
```
,
```
TASKS/
```
,
```
SESSIONS/
```
)
Files: ALL-CAPS for critical files (
```
README.md
```
,
```
RULES.md
```
), kebab-case for others

顶级目录：全部大写（
```
SYSTEM/
```
、
```
TASKS/
```
、
```
SESSIONS/
```
）
文件：关键文件全部大写（
```
README.md
```
、
```
RULES.md
```
），其他文件使用短横线分隔命名（kebab-case）

Session Files

会话文件

One file per day:
```
YYYY-MM-DD.md
```
Multiple sessions same day use Session 1, Session 2, etc. in the same file

每日一个文件：
```
YYYY-MM-DD.md
```
同一天的多个会话在同一文件中使用Session 1、Session 2等区分

Customization

自定义配置

After scaffolding, customize:

```
SYSTEM/PRD.md
```
- Fill in your product requirements (use with fullstack-workspace-init)
```
SYSTEM/ENTITIES.md
```
- Document your data entities
```
SYSTEM/RULES.md
```
- Add project-specific coding standards
```
SYSTEM/ARCHITECTURE.md
```
- Document your architecture
```
SYSTEM/critical/CRITICAL-NEVER-DO.md
```
- Add project-specific violations
```
SOP/
```
- Add your standard operating procedures
```
.claude/rules/
```
- Add project-specific rules
```
.claude/commands/
```
- Add project-specific commands

搭建完成后，可进行以下自定义：

```
SYSTEM/PRD.md
```
- 填写产品需求（与fullstack-workspace-init配合使用）
```
SYSTEM/ENTITIES.md
```
- 记录数据实体
```
SYSTEM/RULES.md
```
- 添加项目专属编码标准
```
SYSTEM/ARCHITECTURE.md
```
- 记录架构信息
```
SYSTEM/critical/CRITICAL-NEVER-DO.md
```
- 添加项目专属禁忌事项
```
SOP/
```
- 添加标准操作流程
```
.claude/rules/
```
- 添加项目专属规则
```
.claude/commands/
```
- 添加项目专属命令

Integration with Other Skills

与其他技能的集成

This skill integrates with:

Skill	How It Works Together
`fullstack-workspace-init`	Use PRD.md to define requirements before scaffolding
`linter-formatter-init`	Sets up quality tooling in the scaffolded project
`husky-test-coverage`	Enforces test coverage in pre-commit hooks

此技能可与以下技能集成：

技能	协作方式
`fullstack-workspace-init`	在搭建前使用PRD.md定义需求
`linter-formatter-init`	在搭建好的项目中配置质量工具
`husky-test-coverage`	在提交前钩子中强制执行测试覆盖率