Token导航 LogoToken导航TokenDH.com
开发敏感数据github未标认证来源可访问许可证需确认审计提醒

claude-plugin-creatorClaude plugin creator 命令行

Agent Skill

claude-plugin-creator 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。

总安装

250

周安装

10

GitHub Stars

公开资料未说明

下载量

81
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

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:claude-plugin-creator(Claude plugin creator 命令行)
来源仓库:https://github.com/bonkey/skills
仓库路径:skills/claude-plugin-creator
安装命令:
npx skills add https://github.com/bonkey/skills --skill claude-plugin-creator
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/bonkey/skills --skill claude-plugin-creator

简介

用于创建和管理 Claude Code 插件的开发指南。

  • 支持判断插件类型(技能/命令)及是否需要外部服务连接。
  • 通过 npx 安装,需根据项目共享需求选择合适架构。
  • 建议参考官方插件规范,确保兼容性和安全性设计。
  • claude-plugin-creator 属于开发类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Claude Plugin Creator

Guide for creating Claude Code plugins. For full reference, see references/plugins-reference.md.

Step 1: Decide What to Build

Ask the user these questions to determine the right approach:

Q1: Will this be used in one project or shared across projects/team?

  • One project → standalone (.claude/ directory) or skill
  • Shared → plugin

Q2: Should Claude use it automatically or should the user invoke it explicitly?

  • Automatically (context-based) → skill (SKILL.md)
  • Explicitly (/command) → command (markdown file)

Q3: Does it need to connect to external services/APIs?

  • Yes → MCP server (standalone or inside plugin)
  • No → skill or command

Q4: Should it run code automatically on events (file save, tool use, etc.)?

  • Yes → hook
  • No → skill or command

Q5: Does it need a specialized AI agent with its own system prompt?

  • Yes → agent
  • No → skill or command

Decision Summary

NeedSolutionLocation
Personal, one project, auto-triggeredStandalone skill.claude/skills/<name>/SKILL.md
Personal, one project, user-invokedStandalone command.claude/commands/<name>.md
Shared, auto-triggeredPlugin with skillplugin/skills/<name>/SKILL.md
Shared, user-invokedPlugin with commandplugin/commands/<name>.md
External API accessMCP server.mcp.json (standalone or plugin)
Event automationHookhooks/hooks.json (standalone or plugin)
Specialized AI roleAgentagents/<name>.md
Multiple of the abovePlugin combining componentsSee plugin structure below

Scope: Who Can Access It?

ScopeFileVisibility
user~/.claude/settings.jsonYou only, all projects
project.claude/settings.jsonTeam via git, this project
local.claude/settings.local.jsonYou only, this project (gitignored)
managedAdmin-controlledOrg-wide, read-only

Step 2: Create the Plugin

Directory Structure

my-plugin/
├── .claude-plugin/
│   └── plugin.json           # Only this file goes here
├── skills/                   # Auto-triggered by context
│   └── skill-name/
│       └── SKILL.md
├── commands/                 # User-invoked via /plugin:command
│   └── command-name.md
├── agents/                   # Specialized subagents
│   └── agent-name.md
├── hooks/
│   └── hooks.json            # Event handlers
├── bin/                      # Executables added to PATH
├── settings.json             # Default settings
├── .mcp.json                 # MCP server configs
└── .lsp.json                 # LSP server configs

Critical rules:

  • Components go at plugin root, NOT inside .claude-plugin/
  • Plugin name must be kebab-case
  • Skills must use SKILL.md (exact case)

plugin.json (minimal)

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

plugin.json (full)

{
  "name": "my-plugin",
  "version": "1.0.0",
  "description": "What the plugin does",
  "author": { "name": "Name", "email": "email@example.com" },
  "homepage": "https://docs.example.com",
  "repository": "https://github.com/user/plugin",
  "license": "MIT",
  "keywords": ["keyword1"],
  "userConfig": {
    "api_key": { "description": "API key", "sensitive": true }
  }
}

User config values available as ${user_config.KEY} in configs and CLAUDE_PLUGIN_OPTION_<KEY> env vars.

Skill (SKILL.md)

---
name: skill-name
description: What it does. Use when [trigger phrases].
---

Instructions for Claude.

Command (commands/name.md)

---
description: What it does
disable-model-invocation: true
---

Instructions. User input: $ARGUMENTS

Agent (agents/name.md)

---
name: agent-name
description: What it specializes in
model: sonnet
maxTurns: 20
---

System prompt here.

Hook (hooks/hooks.json)

{
  "hooks": {
    "PostToolUse": [{
      "matcher": "Write|Edit",
      "hooks": [{ "type": "command", "command": "${CLAUDE_PLUGIN_ROOT}/scripts/lint.sh" }]
    }]
  }
}

MCP Server (.mcp.json)

{
  "mcpServers": {
    "server-name": {
      "command": "${CLAUDE_PLUGIN_ROOT}/servers/server",
      "args": ["--config", "${CLAUDE_PLUGIN_ROOT}/config.json"]
    }
  }
}

Step 3: Test

claude --plugin-dir ./my-plugin

Inside session: /reload-plugins after changes. Validate: claude plugin validate.

Step 4: Distribute

Option A: Marketplace (for team/community)

Create .claude-plugin/marketplace.json in marketplace repo:

{
  "name": "marketplace-name",
  "owner": { "name": "Team Name" },
  "plugins": [
    {
      "name": "plugin-name",
      "source": "./plugins/plugin-name",
      "description": "What it does",
      "version": "1.0.0"
    }
  ]
}

Plugin sources can be: relative path ("./plugins/x"), GitHub ({"source": "github", "repo": "owner/repo"}), git URL, git subdirectory, or npm package.

Host on GitHub, add with: /plugin marketplace add owner/repo

Option B: Direct sharing

Host plugin directory on GitHub. Users install with --plugin-dir or add to their .claude/ config.

Option C: Team auto-install

Add to project .claude/settings.json:

{
  "extraKnownMarketplaces": {
    "team-tools": { "source": { "source": "github", "repo": "org/plugins" } }
  },
  "enabledPlugins": { "my-plugin@team-tools": true }
}

Option D: Official marketplace

Submit at claude.ai/settings/plugins/submit or platform.claude.com/plugins/submit.

Converting Standalone to Plugin

  1. mkdir -p my-plugin/.claude-plugin
  2. Create plugin.json with name
  3. Copy .claude/commands/my-plugin/commands/
  4. Copy .claude/skills/my-plugin/skills/
  5. Copy .claude/agents/my-plugin/agents/
  6. Move hooks from settings.json → my-plugin/hooks/hooks.json
  7. Test: claude --plugin-dir./my-plugin

Environment Variables in Plugins

  • ${CLAUDE_PLUGIN_ROOT} — plugin install path (changes on update)
  • ${CLAUDE_PLUGIN_DATA} — persistent data path (survives updates)

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

36.56%
按下载量换算30

Claude

27.62%
按下载量换算22

Cursor

17.4%
按下载量换算14

Gemini CLI

8.96%
按下载量换算7

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

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。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills