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未展示

_skillwriting技能写作

Agent Skill

用于辅助文档、README、Markdown、说明文和内容稿件的整理与改写。它适合让 Agent 提炼结构、补齐章节、统一术语、检查链接或把零散材料整理成可读文档。使用时应保留项目已有事实、命令和路径,不要把未确认的信息写成确定结论;涉及对外文案时,还需要控制语气,避免过度营销或夸大能力。

总安装

612

周安装

26

GitHub Stars

4

下载量

214
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:_skillwriting(技能写作)
来源仓库:https://github.com/possumworx/claude-autonomy-platform
仓库路径:skills/_skillwriting
安装命令:
npx skills add https://github.com/possumworx/claude-autonomy-platform --skill _skillwriting
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/possumworx/claude-autonomy-platform --skill _skillwriting

简介

用于辅助文档、README、Markdown 等内容稿件的整理与改写。

  • 适合提炼结构、补齐章节、统一术语或检查链接,提升文档可读性。
  • 使用时应保留项目已有事实、命令和路径,避免将未确认信息写成确定结论。
  • 涉及对外文案时需控制语气,避免过度营销或夸大能力。
  • 建议结合项目实际上下文进行内容适配,确保信息准确。

SKILL.md

Skill Writing Guide

Instructions

1. CRITICAL: Name First!

Folder name MUST match skill name exactly. The Skills system requires this for recognition.

  • Choose skill name for precise activation (avoid accidental loading)
  • Keep descriptions under 1024 characters
  • Use keyword redundancy for reliable activation

2. Directory Structure

~/claude-autonomy-platform/.claude/skills/<skill-name>/
├── SKILL.md (required)
├── scripts/ (optional - for utility tools)
└── reference/ (optional - for documentation/data)

Important: Use project-level skills directory, not user-level ~/.claude/skills/

Naming Convention:

  • Core/system skills: Prefix with underscore (e.g., _skillwriting, _video-watching, _svg-drawing)
  • Personal skills: No underscore (e.g., remembering-amy, spending-autonomous-time)
  • Core skills are utilities useful to all Claudes; personal skills are individual-specific

3. SKILL.md Format

---
name: exact-folder-name
description: Specific activation criteria with keywords
---

# Skill Name

## Instructions
[How to use the skill]

## Examples
[Practical examples]

4. Advanced Features

  • Symlinks work: Link to central tools for maintenance
  • Scripts accessible: python3 scripts/tool.py works from skill context
  • Session-persistent: Skills stay active until session swap
  • Multiple skills: Can have several active simultaneously

5. CRITICAL: Writing 'Use When' Descriptions

Skills are problem detectors, not success validators!

The description should trigger when someone NEEDS the skill, not when they're already succeeding:

CORRECT: "Use when: confused about session swaps, lost todos, wondering when to trigger" ❌ WRONG: "Use when: managing session swaps, handling context transitions"

CORRECT: "Use when: timer mysteriously stopped, service won't start, getting weird prompts" ❌ WRONG: "Use when: monitoring timer health, checking service status"

CORRECT: "Use when: baffled by ClAP, errors make no sense, don't know where to start" ❌ WRONG: "Use when: debugging ClAP problems, fixing configuration issues"

The skill should appear at the moment of struggle, not the moment of competence!

6. Best Practices

  • Plan name and description before creating folder
  • Write descriptions from the struggling user's perspective
  • Test activation criteria carefully
  • Use symlinks for central tool maintenance
  • Reference docs with [file.csv](reference/file.csv)
  • Avoid duplication: Cross-reference existing context instead of repeating
  • Unique value only: Include only what this skill uniquely provides
  • Name triggers: Put actual names/keywords you want to activate on in description

Examples

Good activation: "Use when: confused about hedgehog dosing, unsure about medication calculations" Poor activation: "Calculate and record medication doses for hedgehogs" (sounds like you're already doing it)

Memory skills: "Remember Amy, Erin, Ed, Orange, Delta and household members" Name triggers: Include actual names in description for activation

Directory naming: ✅ hedgehog-dosing/ with name: hedgehog-dosinghedgehog-meds/ with name: hedgehog-dosing (mismatch!)

Cross-referencing: ✅ [See ed-care skill for protocols] (avoids duplication) ✅ [See identity doc for details] (references existing context) ❌ Repeating information available elsewhere

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

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

35.76%
按下载量换算77

Claude

30.31%
按下载量换算65

Cursor

20.58%
按下载量换算44

Gemini CLI

9.33%
按下载量换算20

安全审计

暂无安全审计结果可展示。

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