Token导航 LogoToken导航

Agent Skill · AI 工具

project-init-orchestrator(项目初始化协调器)

project-init-orchestrator 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态、代码变更或协作事项进行整理时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

project-init-orchestrator 是AI 工具类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为需要联网;许可证为 MIT,第三方扫描已通过,按安装量在AI 工具类的 2,536 个 Skill 里排第 693。

总安装 2,446周安装 104GitHub Stars 21下载 857核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:project-init-orchestrator(项目初始化协调器)
来源仓库:https://github.com/shipshitdev/library
仓库路径:skills/project-init-orchestrator
安装命令:
npx skills add https://github.com/shipshitdev/library --skill project-init-orchestrator
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/shipshitdev/library --skill project-init-orchestrator

这个 Skill 是做什么的

AI 工具类 Skill 用来扩展助手本身的能力,例如接模型、接检索、接外部服务。

project-init-orchestrator 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中围绕仓库状态、代码变更或协作事项进行整理。

适用于项目初始化协调、多步骤流程管理和协作事项调度等 AI 工具类任务。可结合来源仓库和安装命令进一步核验具体用法和功能边界。安装前建议确认权限范围和维护状态,避免触发不必要的联网或命令执行。需注意是否会读写文件或访问私有仓库,确保 token 权限和用户授权合规。

来源仓库内路径:skills/project-init-orchestrator

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/shipshitdev/library --skill project-init-orche…;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要需要联网,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

project-init-orchestrator 怎么安装?

在终端执行 npx skills add https://github.com/shipshitdev/library --skill project-init-orchestrator。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

project-init-orchestrator 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

project-init-orchestrator 需要什么权限?

权限级别是需要联网。该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

project-init-orchestrator 是免费的吗?

来源标注的许可证是 MIT。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

project-init-orchestrator 用的人多吗?

公开数据是累计安装 2,446 次,近一周安装 104 次,下载 857 次,GitHub 21 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

project-init-orchestrator 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 3 个目录同时收录,说明这个 Skill 流传比较广。
  • 许可证明确为 MIT,商用和分发边界好判断。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理AI 工具类任务、又不想每次都手写提示词的人。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 需要厂商级技术支持和 SLA 的生产系统。

AI 工具类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
executing-plans

执行计划 · Codex、Claude

1,515,336173,110只读MIT通过
verification-before-completion

完成前的验证 · Codex、Claude

1,319,472173,121权限未标注MIT通过
subagent-driven-development

子 Agent 驱动开发 · Codex、Claude

1,312,416173,138只读MIT通过
copy-editing

复制编辑 · Codex、Claude

1,209,57625,632只读MIT通过
paywall-upgrade-cro

付费专区升级 cro · Codex、Claude

1,030,17625,617只读MIT通过
project-init-orchestrator

项目初始化协调器 · 本页

2,44621需要联网MIT通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Project Init Orchestrator

Overview

This skill orchestrates multiple initialization skills to set up a complete, production-ready project environment. Instead of manually invoking each skill, this orchestrator coordinates them in the correct sequence with proper dependencies.

When to Use This Skill

This skill activates automatically when users:

  • Start a new project from scratch
  • Want full project setup with one command
  • Need AI-first development infrastructure + code quality tools
  • Say "initialize project", "set up new project", "bootstrap project"
  • Want consistent setup across multiple projects

Skills Orchestrated

OrderSkillPurposeRequired
1agent-folder-initAI documentation & standardsYes
2linter-formatter-initESLint + Prettier + pre-commitYes
3husky-test-coverageTest coverage enforcementOptional
4Component scaffoldingBackend/Frontend/Mobile/ExtensionOptional

Orchestration Flow

┌─────────────────────────────────────────────────────────────┐
│              PROJECT INIT ORCHESTRATOR                       │
└─────────────────────────────────────────────────────────────┘
                           │
                           ▼
┌─────────────────────────────────────────────────────────────┐
│  PHASE 1: GATHER CONTEXT                                    │
│  • Project name and path                                    │
│  • Tech stack (Next.js, NestJS, Expo, Plasmo)              │
│  • Package manager preference (bun, pnpm, npm)             │
│  • Test coverage threshold (default: 80%)                  │
│  • Additional scaffolding needs                            │
└─────────────────────────────────────────────────────────────┘
                           │
                           ▼
┌─────────────────────────────────────────────────────────────┐
│  PHASE 2: AGENT FOLDER INIT                                 │
│  • Create .agents/ directory structure                       │
│  • Set up SESSIONS/, TASKS/, SYSTEM/ folders               │
│  • Generate coding standards and rules                      │
│  • Copy agent configs (.claude/, .codex/, .cursor/)        │
│  ──────────────────────────────────────────────────────────│
│  Invocation:                                                │
│  python3 scripts/scaffold.py                               │
│          (from agent-folder-init skill)                    │
│          --root /path/to/project                           │
└─────────────────────────────────────────────────────────────┘
                           │
                           ▼
┌─────────────────────────────────────────────────────────────┐
│  PHASE 3: LINTER FORMATTER INIT                             │
│  • Detect project tech stack                                │
│  • Install ESLint + Prettier (or Biome)                    │
│  • Configure framework-specific rules                       │
│  • Set up lint-staged for pre-commit                       │
│  • Create .vscode/settings.json                            │
│  ──────────────────────────────────────────────────────────│
│  Invocation:                                                │
│  Use linter-formatter-init skill guidance                   │
└─────────────────────────────────────────────────────────────┘
                           │
                           ▼
┌─────────────────────────────────────────────────────────────┐
│  PHASE 4: HUSKY TEST COVERAGE (if tests exist)             │
│  • Detect test runner (Jest, Vitest, Mocha)                │
│  • Configure coverage thresholds                            │
│  • Add pre-commit hook for test coverage                   │
│  ──────────────────────────────────────────────────────────│
│  Invocation:                                                │
│  Use husky-test-coverage skill guidance                     │
└─────────────────────────────────────────────────────────────┘
                           │
                           ▼
┌─────────────────────────────────────────────────────────────┐
│  PHASE 5: COMPONENT SCAFFOLD (optional)                     │
│  • Scaffold additional components if requested:            │
│    - Backend (NestJS + MongoDB + Swagger + Dockerfile)     │
│    - Frontend (Next.js + Tailwind + App Router)            │
│    - Mobile (Expo + Expo Router + React Native)            │
│    - Extension (Plasmo + React + Tailwind)                 │
│  • Supports monorepo (workspaces) or separate repos       │
│  ──────────────────────────────────────────────────────────│
│  Invocation:                                                │
│  python3 scripts/scaffold.py                               │
└─────────────────────────────────────────────────────────────┘
                           │
                           ▼
┌─────────────────────────────────────────────────────────────┐
│  PHASE 6: VERIFICATION                                      │
│  • Verify all configurations created                        │
│  • Run lint check (should pass)                            │
│  • Confirm git hooks installed                             │
│  • Generate setup summary                                   │
└─────────────────────────────────────────────────────────────┘

Usage

Quick Start (Recommended)

When user says "initialize my project" or "set up new project":

1. Ask for project context:
   - Project path (default: current directory)
   - Tech stack (Next.js, NestJS, Node.js, etc.)
   - Package manager (bun, pnpm, npm)
   - Test coverage threshold (default: 80%)
   - Need additional scaffolding? (backend, frontend, mobile, extension)

2. Execute phases in order:
   Phase 2 → Phase 3 → Phase 4 → Phase 5 (if needed) → Phase 6

Manual Orchestration

If you need to run phases individually:

Phase 2: Agent Folder Init

python3 scripts/scaffold.py --root /path/to/project  # from agent-folder-init skill

Phase 3: Linter Formatter Follow the linter-formatter-init skill to:

  • Install dependencies based on detected stack
  • Configure ESLint rules
  • Set up Prettier
  • Configure lint-staged

Phase 4: Test Coverage Follow the husky-test-coverage skill to:

  • Detect test runner
  • Configure coverage thresholds
  • Add pre-commit hook

Phase 5: Component Scaffold (optional)

python3 scripts/scaffold.py

Supports scaffolding:

  • Backend (NestJS): MongoDB, Swagger, soft deletes (isDeleted), multi-tenancy (filter by organization), Dockerfile
  • Frontend (Next.js): Tailwind CSS, TypeScript strict, App Router, path aliases (@components/, @services/, @hooks/)
  • Mobile (Expo): Expo Router, TypeScript, platform-specific configs
  • Extension (Plasmo): React + TypeScript, Tailwind, manifest config, popup component

Structure options:

  • Monorepo: All components in one repo with workspace config
  • Separate repos: Each component in its own directory
  • Existing projects: Add components incrementally

Configuration Presets

Minimal (AI docs + linting)

Phases: 2, 3
Output:
├── .agents/
├── .eslintrc.js
├── .prettierrc
├── .husky/pre-commit (lint-staged)
└── .vscode/settings.json

Standard (+ test coverage)

Phases: 2, 3, 4
Output:
├── .agents/
├── .eslintrc.js
├── .prettierrc
├── .husky/pre-commit (lint-staged + tests)
├── jest.config.js (coverage thresholds)
└── .vscode/settings.json

Full Stack (+ scaffolding)

Phases: 2, 3, 4, 5
Output:
├── .agents/
├── apps/
│   ├── web/          (Next.js)
│   ├── api/          (NestJS)
│   ├── mobile/       (Expo)
│   └── extension/    (Plasmo)
├── packages/
│   └── shared/
├── .eslintrc.js
├── .prettierrc
├── .husky/pre-commit
└── .vscode/settings.json

Generated Structure

After full orchestration, your project will have:

project-root/
├── .agents/                          # AI-first documentation
│   ├── SESSIONS/                    # Daily session logs
│   ├── TASKS/                       # Task tracking
│   │   └── INBOX.md
│   ├── SYSTEM/                      # Architecture docs
│   │   ├── ARCHITECTURE.md
│   │   └── RULES.md
│   └── README.md
│
├── .claude/                         # Claude Code configs
│   ├── commands/
│   ├── rules/
│   └── skills/
│
├── .husky/                          # Git hooks
│   └── pre-commit                   # Runs lint-staged + tests
│
├── .vscode/                         # Editor settings
│   └── settings.json                # Format on save
│
├── .eslintrc.js                     # ESLint configuration
├── .prettierrc                      # Prettier configuration
├── .lintstagedrc                    # Lint-staged configuration
│
└── [app directories if scaffolded]

Best Practices

Sequencing

  • Always run agent-folder-init FIRST (establishes standards)
  • Run linter-formatter-init BEFORE husky-test-coverage (both use Husky)
  • Run component scaffolding LAST (depends on base setup)

Customization

  • Each phase can be skipped if already configured
  • Phases are idempotent (safe to re-run)
  • Configuration can be customized after orchestration

Verification Checklist

After orchestration, verify:

  • .agents/ folder exists with proper structure
  • npm run lint or bun lint works
  • npm run format or bun format works
  • Git commit triggers pre-commit hooks
  • Tests run on commit (if Phase 4 enabled)

Example Interaction

User: "Set up my new project at ~/projects/my-saas"

Orchestrator Response:

I'll initialize your project with full AI-first infrastructure.

Let me gather some context:
1. Tech stack: [Next.js / NestJS / Node.js / Other]
2. Package manager: [bun / pnpm / npm]
3. Test coverage threshold: [80% default]
4. Need scaffolding? [backend / frontend / mobile / extension]

[After user responds]

Starting orchestration...

✓ Phase 2: Agent Folder Init
  - Created .agents/ with 13 files
  - Copied agent configs (.claude/, .codex/, .cursor/)

✓ Phase 3: Linter Formatter Init
  - Detected: Next.js + TypeScript
  - Installed: eslint, prettier, lint-staged
  - Created: .eslintrc.js, .prettierrc, .lintstagedrc

✓ Phase 4: Husky Test Coverage
  - Detected: Jest
  - Set threshold: 80%
  - Added pre-commit hook

✓ Phase 6: Verification
  - All configs created ✓
  - Lint check passed ✓
  - Git hooks installed ✓

Project initialized successfully!

Troubleshooting

Phase fails to complete

  • Check if dependencies are installed (node, python3)
  • Verify write permissions to project directory
  • Check if conflicting configs exist

Linting errors after setup

  • Run npm run lint:fix to auto-fix
  • Check .eslintrc.js matches your stack
  • Verify TypeScript config if using TS

Pre-commit hook not running

  • Run npx husky install manually
  • Check .husky/pre-commit exists and is executable
  • Verify prepare script in package.json

平台分布

Claude Code

28.7%
按下载量换算246

Gemini CLI

22.21%
按下载量换算190

Antigravity

18.03%
按下载量换算155

OpenCode

11.03%
按下载量换算95

Codex

8.34%
按下载量换算71

Cursor

3.62%
按下载量换算31

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。

继续浏览同类 Skills