Token导航 LogoToken导航TokenDH.com
开发操作浏览器github未标认证来源可访问许可证需确认审计通过

technical-writing技术写作

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

1,673

周安装

69

GitHub Stars

1

下载量

546
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:technical-writing(技术写作)
来源仓库:https://github.com/tenzir/skills
仓库路径:skills/technical-writing
安装命令:
npx skills add https://github.com/tenzir/skills --skill technical-writing
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/tenzir/skills --skill technical-writing

简介

用于辅助文档、Markdown 和内容稿件的整理与改写。

  • 可提炼结构、补齐章节或检查链接,提升文档质量。
  • 保留项目已有事实和命令,不将未确认信息写成确定结论。
  • 对外文案需注意语气,避免过度营销或夸大能力。
  • technical-writing 属于开发类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Technical Writing

Write clear, consistent technical documentation based on Google's developer documentation style guide.

Core Principles

  • Clarity over cleverness: Prioritize understandable prose over elegant writing
  • Consistency matters: Apply rules uniformly throughout a document
  • Audience-first: Write for developers with varying English proficiency

Voice and Tone

Use second person: Address the reader directly with "you."

Good: "You can configure the server by editing config.yaml." Bad: "We can configure the server by editing config.yaml."

Use active voice: Make clear who performs the action.

Good: "The function returns an error code." Bad: "An error code is returned by the function."

Be conversational but professional: Write naturally without slang, jargon, or excessive formality. Use common contractions like "you're," "don't," and "it's" to maintain an informal tone.

What to Avoid

  • Jargon and buzzwords: "leverage," "utilize," "synergy"
  • Filler phrases: "please note that," "it should be noted"
  • Exclamation marks: Reserve for genuinely exciting moments
  • Implying simplicity: "simply," "just," "easy" (frustrates struggling readers)
  • Latin abbreviations: Write "for example" not "e.g.", "that is" not "i.e."
  • Directional language: Use "earlier/later" not "above/below"
  • Anthropomorphic language: Don't attribute human qualities to software ("the system wants," "the server thinks")
  • Skipping articles: Always include "a," "an," and "the"—even in headings

Inclusive Language

AvoidUse Instead
blacklist/whitelistdenylist/allowlist, blocklist/safelist
master/slaveprimary/replica, leader/follower
man-hoursperson-hours
guyseveryone, folks, team
sanity checkquick check, coherence check

Capitalization

  • Sentence case for headings: Capitalize only the first word and proper nouns
  • Product names: Always capitalize (Tenzir, Kubernetes, Docker)
  • Acronyms: Use standard capitalization (API, HTTP, JSON)

Headings

  • Task headings: Use bare infinitives ("Create an instance")
  • Concept headings: Use noun phrases ("User authentication")
  • Hierarchy: Don't skip levels (h1 to h3 without h2)
  • No links: Avoid hyperlinks within headings

Lists

Numbered lists: For sequential steps or ordered items.

Bulleted lists: For non-sequential items.

List guidelines:

  • Introduce with a complete sentence followed by a colon
  • Start each item with a capital letter
  • Use parallel structure across items
  • End with periods if items contain verbs or are complete sentences
  • Don't use single-item lists

Procedures

Structure:

  1. Introduce with context
  2. Number each step
  3. Start steps with imperative verbs
  4. State location before action

Guidelines:

  • "In the Settings panel, click Save." (location before action)
  • Mark optional steps with "Optional:" at the start
  • Document one accessible procedure, not multiple alternatives
  • Avoid "please" in instructions

Code in Text

Use code font (backticks) for:

  • Command-line input and output
  • Filenames and paths
  • Class, method, and function names
  • Keywords, variables, and parameters
  • HTTP status codes and methods

Don't use code font for:

  • Product names and services
  • URLs meant to be followed in a browser
  • Domain names

Grammar: Don't inflect code elements. Write "send a POST request" not "POST the data."

Word Choices

AvoidUse Instead
akaalso known as, or use parentheses
allows you tolets you
configconfiguration
adminadministrator
click onclick
check (checkbox)select
uncheckclear
higher/lower (versions)later/earlier

Formatting

  • UI elements: Bold (File > Save)
  • User input: Code font or bold, depending on context
  • New terms: Italics on first use
  • Emphasis: Use sparingly; prefer restructuring sentences

Links

  • Use descriptive link text that makes sense out of context
  • Avoid "click here" or "this page"
  • Good: "See the authentication guide."
  • Bad: "Click here for more information."

Reference Hierarchy

When style questions arise, consult in order:

  1. Project-specific guidelines
  2. Google Developer Documentation Style Guide
  3. Merriam-Webster (spelling)
  4. Chicago Manual of Style (general)
  5. Microsoft Writing Style Guide (technical)

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

31.96%
按下载量换算175

Claude

29.75%
按下载量换算162

Cursor

18.89%
按下载量换算103

Gemini CLI

9.41%
按下载量换算51

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

操作浏览器

该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

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