Token导航 LogoToken导航

Agent Skill · 开发

metaskill-packaging(元技能包装)

metaskill-packaging 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。

metaskill-packaging 是开发类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为调用外部服务;许可证信息还不明确,第三方扫描已通过,按安装量在开发类的 21,746 个 Skill 里排第 21,023。

总安装 196周安装 8GitHub Stars 3下载 63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
unknown

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:metaskill-packaging(元技能包装)
来源仓库:https://github.com/gigaverse-app/skillet
仓库路径:skills/metaskill-packaging
安装命令:
npx skills add https://github.com/gigaverse-app/skillet --skill metaskill-packaging
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。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/gigaverse-app/skillet --skill metaskill-packaging

这个 Skill 是做什么的

开发类 Skill 面向写代码、改代码和排查问题,一般会读写仓库文件,部分还会执行构建或测试命令。

用于处理 GitHub 仓库、Issue、Pull Request 等协作信息。适合围绕代码变更、项目状态或协作事项进行整理和分析。

安装命令:npx skills add https://github.com/gigaverse-app/skillet --skill metaskill-packaging 涉及写入操作时需确认 token 权限和仓库访问范围。建议核对原始 README 了解具体功能边界。

来源仓库内路径:skills/metaskill-packaging

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/gigaverse-app/skillet --skill metaskill-packag…;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

    它需要调用外部服务,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

metaskill-packaging 怎么安装?

在终端执行 npx skills add https://github.com/gigaverse-app/skillet --skill metaskill-packaging。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

metaskill-packaging 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。

metaskill-packaging 需要什么权限?

权限级别是调用外部服务。该 Skill 可能调用第三方服务、云服务或外部模型 API,使用前需要确认账号、额度、数据发送范围和服务条款。

metaskill-packaging 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

metaskill-packaging 用的人多吗?

公开数据是累计安装 196 次,近一周安装 8 次,下载 63 次,GitHub 3 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。

metaskill-packaging 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。
  • 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

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 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。
  • 需要处理开发类任务、又不想每次都手写提示词的人。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

开发类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
skill-vetter

技能审查员 · OpenClaw

8,779,018986操作浏览器MIT-0通过
ontology

本体论 · OpenClaw

6,636,960576只读MIT-0通过
entra-app-registration

内部应用程序注册 · Codex、Claude

6,631,872764接触敏感数据未明确通过
entra-app-registration

内部应用程序注册 · Codex、Claude

3,757,680201接触敏感数据未明确通过
nano-banana-pro

纳米香蕉专业版 · OpenClaw

3,526,898364接触敏感数据MIT-0有提醒
metaskill-packaging

元技能包装 · 本页

1963调用外部服务未明确通过

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

Plugin Building and Packaging

Package your skills, agents, commands, and hooks as distributable plugins.

Plugin Structure

CRITICAL RULE: Only plugin.json goes inside .claude-plugin/. All components go at the plugin ROOT:

my-plugin/                      <- Plugin name = neutral noun
├── .claude-plugin/
│   └── plugin.json             # ONLY this file here!
├── commands/                   # Slash commands (*.md) - imperative verbs
├── agents/                     # Agent definitions (*.md) - role nouns
├── skills/                     # Skills (*/SKILL.md) - ending in -ing
├── hooks/                      # Event handlers (hooks.json)
├── .mcp.json                   # MCP servers (optional)
├── .lsp.json                   # LSP servers (optional)
└── README.md
# ❌ WRONG - components inside .claude-plugin/
.claude-plugin/
├── plugin.json
├── commands/         ← NO!
└── skills/           ← NO!

# ✅ CORRECT - components at plugin root
.claude-plugin/
└── plugin.json
commands/             ← YES!
skills/               ← YES!

plugin.json Manifest

Required fields:

{
  "name": "my-plugin",
  "version": "1.0.0",
  "description": "What this plugin does"
}

With recommended metadata:

{
  "name": "my-plugin",
  "version": "1.0.0",
  "description": "What this plugin does",
  "author": {
    "name": "Your Name",
    "email": "you@example.com"
  },
  "license": "MIT",
  "keywords": ["keyword1", "keyword2"],
  "repository": "https://github.com/user/repo",
  "homepage": "https://github.com/user/repo#readme"
}

See references/plugin-json-schema.md for the complete field reference.

Naming Conventions

See /metaskill-naming for the full naming convention.

Quick reference:

ComponentFormExample
Plugin nameNeutral nounmetaskill, codeforge, datakit
Skills-ing (gerund)metaskill-authoring, metaskill-triggering
AgentsRole nounmetaskill-trigger-tester, metaskill-analyzer
CommandsImperative verb/metaskill-create, /quick-start

Plugin name = Common prefix of all atoms

# ✅ GOOD - neutral noun prefix, correct suffixes
metaskill/
├── skills/
│   ├── metaskill-authoring/     # -ing
│   └── metaskill-triggering/    # -ing
├── agents/
│   └── metaskill-trigger-tester.md  # role noun
└── commands/
    └── quick-start.md           # imperative

# ❌ BAD - verb-form prefix
skill-authoring/
├── skills/
│   └── skill-authoring-trigger/  # prefix already -ing!

No type postfixes:

# ❌ BAD - redundant type postfix
skills/code-review-skill/
agents/tester-agent.md
commands/lint-command.md

# ✅ GOOD - no type postfix
skills/code-reviewing/
agents/tester.md
commands/lint.md

Dogfooding Approaches

Quick Iteration (Active Development)

claude --plugin-dir ./my-plugin
  • Loads plugin immediately
  • Restart Claude Code to pick up changes
  • Best for rapid iteration

Marketplace Testing (Pre-Release)

# Add your repo as a local marketplace (once)
/plugin marketplace add /path/to/your/repo

# Install the plugin
/plugin install your-repo@my-plugin

# After changes, reinstall to test
/plugin uninstall your-repo@my-plugin
/plugin install your-repo@my-plugin
  • Tests the full installation flow
  • Verifies the user experience
  • Use before releasing

Plugin Placement in Repos

Single Plugin at Repo Root

For a repo that IS the plugin:

my-plugin-repo/
├── .claude-plugin/
│   └── plugin.json
├── skills/
│   └── my-plugin-authoring/
├── agents/
└── README.md

Plugin Inside a Project (Dogfooding)

For internal tooling within a larger project:

my-project/
├── src/
├── tests/
├── .claude/                  # Project's Claude config
│   └── settings.json
└── plugins/
    └── my-internal-plugin/
        ├── .claude-plugin/
        │   └── plugin.json
        └── skills/

Load with: claude --plugin-dir./plugins/my-internal-plugin

Multiple Plugins in One Repo

Use marketplace.json to reference multiple plugins:

my-repo/
├── .claude-plugin/
│   └── marketplace.json      # References plugins below
├── plugins/
│   ├── plugin-a/
│   │   ├── .claude-plugin/
│   │   │   └── plugin.json
│   │   └── skills/
│   └── plugin-b/
│       ├── .claude-plugin/
│       │   └── plugin.json
│       └── agents/
└── README.md

marketplace.json (required fields):

{
  "name": "my-marketplace",
  "owner": {
    "name": "Your Name"
  },
  "plugins": [
    { "name": "plugin-a", "source": "./plugins/plugin-a" },
    { "name": "plugin-b", "source": "./plugins/plugin-b" }
  ]
}

With full metadata:

{
  "name": "my-marketplace",
  "owner": {
    "name": "Your Name",
    "email": "you@example.com"
  },
  "metadata": {
    "description": "Description of your marketplace",
    "version": "1.0.0"
  },
  "plugins": [
    {
      "name": "plugin-a",
      "source": "./plugins/plugin-a",
      "description": "What plugin-a does",
      "version": "1.0.0",
      "author": { "name": "Your Name", "email": "you@example.com" },
      "license": "MIT",
      "keywords": ["keyword1", "keyword2"],
      "category": "development"
    }
  ]
}

See references/marketplace-json-schema.md for the complete field reference.

Users can then:

/plugin marketplace add /path/to/my-repo
/plugin install my-marketplace@plugin-a

Internal vs External Plugins

Internal (Dogfooding within Repo)

Place in a plugins/ or tools/ directory:

my-project/
├── plugins/
│   └── internal-tooling/     # For this project only
│       ├── .claude-plugin/
│       │   └── plugin.json
│       └── skills/
  • Not meant for distribution
  • Project-specific utilities
  • Load with --plugin-dir

External (Open Source / Distribution)

Option A: Dedicated plugin repo

my-plugin/                    # Repo IS the plugin
├── .claude-plugin/
│   └── plugin.json
└── skills/

Option B: Plugin marketplace repo

my-plugins/                   # Repo contains multiple plugins
├── .claude-plugin/
│   └── marketplace.json
└── plugins/
    ├── plugin-a/
    └── plugin-b/

Plugin Components Reference

DirectoryContentsNaming Pattern
.claude-plugin/plugin.json onlyN/A
skills/*/SKILL.mdprefix-action-ing
agents/*.mdprefix-role-noun
commands/*.mdimperative-verb
hooks/hooks.jsonN/A
.mcp.jsonMCP serversN/A
.lsp.jsonLSP serversN/A

Common Mistakes

Components in Wrong Location

# ❌ WRONG
.claude-plugin/
├── plugin.json
└── skills/          # NO! Skills outside .claude-plugin/

# ✅ CORRECT
.claude-plugin/
└── plugin.json
skills/              # YES! At plugin root

Missing plugin.json

# ❌ WRONG - no manifest
my-plugin/
└── skills/

# ✅ CORRECT - has manifest
my-plugin/
├── .claude-plugin/
│   └── plugin.json
└── skills/

Verb-Form Prefix

# ❌ WRONG - prefix is already -ing
skill-authoring/
├── skills/
│   └── skill-authoring-triggering/  # Double verb!

# ✅ CORRECT - neutral noun prefix
metaskill/
├── skills/
│   └── metaskill-triggering/        # Noun + -ing

Related Skills

  • For naming conventions, see /metaskill-naming
  • For skill structure and writing, see /metaskill-authoring
  • For skill group patterns, see /metaskill-grouping
  • For trigger optimization, see /metaskill-triggering
  • To test if triggers work, use the metaskill-trigger-tester agent

平台分布

Codex

34.63%
按下载量换算22

Claude

34.07%
按下载量换算21

Cursor

18.45%
按下载量换算12

Gemini CLI

9.65%
按下载量换算6

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

调用外部服务

该 Skill 可能调用第三方服务、云服务或外部模型 API,使用前需要确认账号、额度、数据发送范围和服务条款。

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。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills