Token导航 LogoToken导航TokenDH.com
待分类只读github未标认证来源可访问许可证需确认审计通过

skill-creation技能创造

Agent Skill

skill-creation 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

729

周安装

31

GitHub Stars

公开资料未说明

下载量

255
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:skill-creation(技能创造)
来源仓库:https://github.com/naykel76/skills-and-guidelines
仓库路径:skills/skill-creation
安装命令:
npx skills add https://github.com/naykel76/skills-and-guidelines --skill skill-creation
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/naykel76/skills-and-guidelines --skill skill-creation

简介

skill-creation 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 中围绕仓库状态进行整理时使用。

  • 适用于待分类任务,可协助分析代码变更、协作事项及项目进展。
  • 通过 npx skills add 命令从指定 GitHub 仓库安装,具体路径为 skills/skill-creation。
  • 安装前需确认权限范围、维护状态,以及是否涉及联网、命令执行或文件读写。
  • 建议结合原始 README 文档进一步核验功能细节与使用边界。

SKILL.md

Frontmatter

  • name — kebab-case, matches the folder name
  • description — trigger mechanism, not a summary

- Always use YAML block scalar (>-) — never write the description as an inline string - Write as "Use this skill whenever [condition]..." - End with "Do not wait for an explicit request — if [condition], this skill applies." unless the skill is only valid when explicitly triggered by the user - No implementation notes or process guidance — trigger conditions only

Body

  • Write directives, not descriptions — tell the agent what to do, not what the skill is about
  • Keep routing, trigger conditions, and non-negotiable constraints in SKILL.md; put subtopic-specific instructions in rules/ files
  • If external docs are referenced, name them explicitly
  • Keep under 500 lines — split large reference content into separate files

Rules Directory

Use a rules/ subdirectory when a skill covers multiple distinct sub-topics that would otherwise bloat the main SKILL.md.

  • Each file in rules/ should cover one focused topic
  • Name files in kebab-case matching the sub-topic (e.g. file-uploads.md)
  • Do not reference a rule file until it exists, or create it in the same change
  • Do not load all rule files by default — instruct the agent to read only the relevant file(s) based on the current task
  • Reference rule files explicitly in SKILL.md using a routing section

Quick Reference section format

Add a ## Quick Reference section to SKILL.md that maps each sub-topic to its rule file with a short summary of key directives:

## Quick Reference

### File Uploads → `rules/file-uploads.md`

- Add upload handling to the component, not the form object
- Configure storage in the form object

### Form Components → `rules/form-component.md`

- Read resource config before building
- Render buttons from config

### Index Components → `rules/index-component.md`

- Build table columns from config
- Render row actions from config

When to use rules vs separate skills

  • Use rules/ when sub-topics share the same trigger condition and would always be activated by the same kind of task
  • Keep separate skills when each sub-topic has a meaningfully different trigger condition that benefits from its own description field

Example

---
name: skill-name
description: >-
  Use this skill whenever [condition]. Do not wait for an explicit
  request — if [condition], this skill applies.
---

## Section

- Directive one
- Directive two

## Quick Reference

### Topic One → `rules/topic-one.md`

- Directive one
- Directive two

### Topic Two → `rules/topic-two.md`

- Directive three
- Directive four

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

36.71%
按下载量换算94

Claude

28.59%
按下载量换算73

Cursor

19.85%
按下载量换算51

Gemini CLI

10.24%
按下载量换算26

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

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。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