Back to directory

Blog Writing Zh

Agent skill
1 skill

Write or rewrite Traditional Chinese (Taiwan) blog posts with a genuine human voice, modeled on seven studied blogs (知識倉鼠, 保哥, 高見龍, 90s.pm.investing, AI避坑情報員, Julia Evans, Simon Willison). Use when asked to 寫部落格文章/ 電子報/blog post, 把筆記改寫成文章, 把 Obsidian 筆記變成 blog, 翻譯改寫外文文章/演講/ 討論串成中文長文, or "用我的風格寫一篇 X", or 不確定該用什麼風格寫(本 skill 會依讀者與目的推薦配方). Supports two modes — compose (from a topic or loose material) and rewrite (from an Obsidian note, translation source, or draft) — and three technical-description modes (操作型可重現教學, 概念型心智 模型導讀, 推演型原理解說) with actionability and concept-clarity checklists — plus composable voice axes (opening strategy × persona intensity × metaphor density × closing move) with presets, an optional dual-draft mode (write two deliberately divergent drafts, then merge the strengths) for long or high-stakes pieces, a selectable length tier (短打/標準/深文/工具書級), and an automatic post-draft check that suggests splitting into a series (with outline) when one piece can't hold the material. Output is always the article PLUS 3-5 title/subtitle candidates. Pipeline position: this skill supplies structure and voice (加法), then ACTIVELY INVOKES avoid-ai-writing-zh as the de-AI finishing pass (減法) — loading and running it (rewrite mode, voice profile mapped from the chosen 風味) rather than merely telling the user to, in any environment that can load sibling skills — plus avoid-china-writing if PRC usage may have leaked in. Do NOT invoke for 正式公文/簽呈 (use formal-doc-structure), RFP (use rfp-writing), 白話翻譯單一術語 (use plain-speak), or pure de-AI editing of an existing text with no restructuring (use avoid-ai-writing-zh directly).

Install with
Claude CodeCodexCursorGitHub CopilotGemini CLI

Supported by the Atlas install prompt. Skill-specific requirements vary.

Skill stats

Skills1CategoryContent & writingStars0

About this skill

Write or rewrite Traditional Chinese (Taiwan) blog posts with a genuine human voice, modeled on seven studied blogs (知識倉鼠, 保哥, 高見龍, 90s.pm.investing, AI避坑情報員, Julia Evans, Simon Willison). Use when asked to 寫部落格文章/ 電子報/blog post, 把筆記改寫成文章, 把 Obsidian 筆記變成 blog, 翻譯改寫外文文章/演講/ 討論串成中文長文, or "用我的風格寫一篇 X", or 不確定該用什麼風格寫(本 skill 會依讀者與目的推薦配方). Supports two modes — compose (from a topic or loose material) and rewrite (from an Obsidian note, translation source, or draft) — and three technical-description modes (操作型可重現教學, 概念型心智 模型導讀, 推演型原理解說) with actionability and concept-clarity checklists — plus composable voice axes (opening strategy × persona intensity × metaphor density × closing move) with presets, an optional dual-draft mode (write two deliberately divergent drafts, then merge the strengths) for long or high-stakes pieces, a selectable length tier (短打/標準/深文/工具書級), and an automatic post-draft check that suggests splitting into a series (with outline) when one piece can't hold the material. Output is always the article PLUS 3-5 title/subtitle candidates. Pipeline position: this skill supplies structure and voice (加法), then ACTIVELY INVOKES avoid-ai-writing-zh as the de-AI finishing pass (減法) — loading and running it (rewrite mode, voice profile mapped from the chosen 風味) rather than merely telling the user to, in any environment that can load sibling skills — plus avoid-china-writing if PRC usage may have leaked in. Do NOT invoke for 正式公文/簽呈 (use formal-doc-structure), RFP (use rfp-writing), 白話翻譯單一術語 (use plain-speak), or pure de-AI editing of an existing text with no restructuring (use avoid-ai-writing-zh directly).

Quick start

  1. Copy the install prompt. Use the green button above this README.
  2. Send it to your agent. Paste it into your coding agent’s chat and send.
  3. Review the setup. Your agent checks the skill’s instructions, requirements, and installation scope.
Review, license & listing details
Review the source before using this skill

This is an imported listing. We haven’t certified its safety, license, or compatibility. Skills may include executable scripts.

Automated security scan: Partial

Only SKILL.md was scanned. Supporting files and scripts were not checked. No safety grade has been assigned.

0 potential findings · Scanned 11/4/2026 · Scanner 0.2.1

Content SHA-256: a24682725e94dfd93af7709773a53dede1c25811df43fefb454f6988e1ce043a

About scan coverage and limitations

About this listing

Category
Content & writing
Reported license
MIT
Listing language (estimated)
Chinese
Review status
Not individually reviewed
Agent compatibility
Atlas installation supports Claude Code, Codex, and Cursor. Skill-specific dependencies and features are checked during setup.
Blog Writing Zh | Agent Skills Atlas