No red flags found in any of the 11 categories — no credential harvesting, no data exfiltration, no curl-pipe-shell installer. Scanned against the SlowMist agent-security taxonomy, refreshed every 8 hours. Full audit →
by Raymondhou0917 · Codex Skill · ★ 960
Last updated: · Indexed by AgentSkillsHub · Auto-synced every 8h
🔒 Is speak-human-tw safe to install? View the security audit →
說人話 speak-human-tw 字都對，卻不像人講的話？那就是 AI 味。 🤖 AI 幫你把稿子寫好了，你自己讀一遍卻覺得哪裡怪，但又說不出來哪裡怪？ 📮 電子報資訊全對，讀起來卻像新聞稿，不像你在跟讀者講話？ 😮💨 每次都要一句一句抓「這句好 AI 味」，叫它重寫，改到最後比自己從頭寫還累？ ✨ 這些， 都能解決。 一個專門校對繁體中文的 skill：抓出 35+ 種 AI 中文寫作痕跡，順手把中國用語和半形標點改成台灣的寫法，改寫成讀起來像人寫的版本，還附一份交稿前能自己打分數的檢核表。 給 Claude Code、Codex、Cursor 和任何能讀 Markdown 的 AI agent 用。 內容創作 · 行銷文案 · 日常辦公 —— 電子報 · 社群貼文 · 銷售頁 · 客服回信 · 簡報 · 公告 · Email 先保住事實 → 再清掉 AI 味 → 最後才加人味 不是把你的稿子洗成機器人，是把 AI 味洗掉、把你洗回來。 說人話，能把你的 AI 味，改成什麼樣？ 社群貼文：emoji 轟炸＋勸誡收尾 改寫前 🚀 重磅消息！AI 寫作工作坊正式上線啦！💡 這不僅僅是一堂課，更是一場思維的革命。與其在原地焦慮，不如現在就出發。畢竟，機會永遠留給準備好的人，不是嗎？🔥 改寫後 AI 寫作工作坊開放報名了。三個小時，教你三件事：怎麼讓 AI 抓到你的語氣、怎麼建自己的改稿流程、怎麼把重複的產文工作自動化。上一梯 40 個名額，三天就額滿，這梯加開到 6
| Stars | 960 |
| Forks | 105 |
| Language | Python |
| Category | Codex Skill |
| License | MIT |
| Quality Score | 59.4409628824209/100 |
| Open Issues | 2 |
| Last Updated | 2026-09-11 |
| Created | 2026-07-08 |
| Platforms | claude-code, codex, python |
| Est. Tokens | ~15k |
These tools work well together with speak-human-tw for enhanced workflows:
Looking for a speak-human-tw alternative? If you're comparing speak-human-tw with other codex skill tools, these 6 projects are the closest alternatives on Agent Skills Hub — ranked by topic overlap, star count, and community traction.
AI-writing humanizer for KO/EN/ZH/JA
去 AI 味：去除简体中文 AI 写作痕迹 / Chinese humanizer skill
说人话｜中文优先的去 AI 味改写 skill：保事实、分场景、改完可直接发。Chinese-first rewrite skill for Codex / Claude Code / Cursor / ChatGPT
A collection of AI agent skills for Clawdbot, Claude Code, Codex
Free, open-source AI writing humanizer and detector. 55 patterns, 5 voices, a 0-100 AI-tell score, and nothing
Universal Claude Code workflow plugin with agents, skills, hooks, and commands
Explore other popular codex skill tools:
speak-human-tw is 「說人話」：繁體中文的去 AI 味改寫 skill。抓 38 種 AI 寫作痕跡，順手校正中國用語與半形標點，給 Claude Code / Codex / Cursor 用。. It is categorized as a Codex Skill with 960 GitHub stars.
speak-human-tw is primarily written in Python. It covers topics such as agent-skills, ai-writing, claude-code.
You can find installation instructions and usage details in the speak-human-tw GitHub repository at github.com/Raymondhou0917/speak-human-tw. The project has 960 stars and 105 forks, indicating an active community.
speak-human-tw is released under the MIT license, making it free to use and modify according to the license terms.
The top alternatives to speak-human-tw on Agent Skills Hub include patina, qu-ai-wei, shuorenhua. Each offers a different approach to the same problem space — compare them side-by-side by stars, quality score, and community activity.
Grades come from a rule-based scan built on the SlowMist agent-security taxonomy, covering 11 red-flag categories including credential harvesting, data exfiltration, and curl | sh installers. It is a first-layer scan, not a manual audit — we say so rather than overstate it.
The scale of the problem is documented independently: Liu et al. (2026), in a study of 31,132 agent skills, report that 26.1% contain security vulnerabilities. Our own full-catalog census is published as a citable open dataset.
Sources & who's responsible: