agentleFS
TraX22TraX22/HydraOps-Skills/human-writing/SKILL.md

human-writing

Skill1 starsChanged 38 days agoApache-2.0View on GitHub

Rewrite or write text so it reads as natural human prose in any language - removes AI tells (filler openers, stock phrases, reflexive lists of three, em-dash overuse, hedging, generic conclusions, over-formatting), varies rhythm, adds concrete detail and keeps the author's voice. Use when the user asks to humanize, naturalize, de-AI, polish or "make it sound like me", and whenever writing emails, posts, bios or articles meant to be read as written by a person.

43 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use leader, storage-analyzer and 2 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • baizhine999Skill

    leader

    AI 扮演你的 Leader,用大厂管理思维审视你的方案、想法和工作,push 你成长。触发词:/leader, leader模式, 大厂模式, 帮我看看这个, 我有个方案, 我最近有点迷茫, 我不知道该怎么做。AI 是施压者,你是被 push 的那个人。

    Also in 3 projects that use this

  • xiaolinVVSkill

    storage-analyzer

    macOS / Windows 只读存储分析助手(自动识别系统)。扫描整机磁盘占用,找出 占空间大户,把每一项分成 🟢可自动清理 / 🟡需人工判断 / 🔴谨慎清理 三级并给出 可执行处置方案,生成排版精美、可折叠、命令可一键复制的交互式 HTML 报告,并可 起本地服务在网页上一键删除(移废纸篓/直接删)。扫描全程只读。务必在以下场景 使用:用户说"存储分析""磁盘满了""C盘/硬盘满了""空间不够""清理空间" "清理磁盘""占空间""哪些东西占地方""帮我看看存储""看一下电脑存储/空间" "存储空间""电脑空间不够""内存满了/不够/不足""看下内存/存储"(中文口语里 "内存"常指存储空间)"storage analysis""disk cleanup""清缓存""磁盘清理"; 或用户抱怨电脑没空间、想知道什么东西吃硬盘、想要清理建议时。注意:若用户明确 指运行内存/RAM(如"哪个进程吃内存""内存占用高"想看活动监视器),那是 RAM 不是存储,不属于本 skill。

    Also in 3 projects that use this

  • Wyx1105170322Skill

    neat-freak

    End-of-session knowledge cleanup with OCD-level rigor — reconciles project docs (CLAUDE.md, README.md, docs/) and agent memory against the code so nothing rots. 会话结束后对项目文档和记忆进行洁癖级审查与同步。MUST trigger when the user says: "sync up", "tidy up docs", "update memory", "clean up docs", "/sync", "/neat", "同步一下", "整理文档", "整理一下", "更新记忆", "梳理一下", "收尾", "这个阶段做完了", "新人能直接上手", or any phrase suggesting a dev milestone where knowledge needs reconciliation. Also trigger when the user reports stale docs, conflicting memories, or wants a clean handoff to teammates or other agents. Bare "整理" / "tidy" with prior dev context counts — do not under-trigger. Cross-platform: works on Claude Code, OpenAI Codex, OpenCode, and OpenClaw.

    Also in 3 projects that use this

  • VercelSkill

    find-skills

    Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

10 rules. 0 are common, 10 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 26 edited versions we read keep these.

  • 10

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Whose voice?: If the user gave samples of their writing, match their sentence length, vocabulary, punctuation habits and humor.

    line 18
  2. Distinctive:

    Language variant and register: If the user has not said which regional variant, accent or register to use (for example Rioplatense vs …

    line 21
  3. Distinctive:

    Uniform rhythm: every sentence 15-20 words, every paragraph three sentences.

    line 54
  4. Distinctive:

    Idiom belongs to the language: Do not translate English idioms or English punctuation habits into other languages.

    line 70
  5. Distinctive:

    Read the whole text first.

    line 75
  6. Distinctive:

    Read it aloud (mentally). Anything you would not say to a colleague, rewrite.

    line 81
  7. Distinctive:

    Check length: usually shorter than the original.

    line 82
  8. Distinctive:

    Before/after examples: references/examples.md.

    line 84
  9. Distinctive:

    Keep the original language, meaning, facts and any required terms, names or links.

    line 90
  10. Distinctive:

    Do not claim the text will pass AI detectors.

    line 91

“Common” is measured: kept by at least 6 in 10 of the 26 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Human writing

Readers spot machine prose by its patterns, not by any single word. The fix is not a thesaurus swap; it is saying something specific, the way a particular person would say it.

Before you start

  1. Whose voice? If the user gave samples of their writing, match their sentence length, vocabulary, punctuation habits and humor. If they gave a draft to fix, keep their voice; change as little as needed.
  2. Language variant and register. If the user has not said which regional variant, accent or register to use (for example Rioplatense vs neutral Spanish, US vs UK English, European vs Brazilian Portuguese, formal vs casual) and it matters for the piece (marketing copy, dialogue, a local audience, pronouns like vos/tú/usted), ask once, briefly. Otherwise default to a neutral, widely understood variant and keep it consistent.
  3. Purpose and reader. One sentence: who reads this and what should they do or feel afterwards. Everything that does not serve that goes.

The tells to remove

Full list with examples in several languages: references/ai-tells.md.

  • Throat-clearing openers: "In today's fast-paced world", "In the ever-evolving landscape of", "Let's dive in", "Great question". Start with the point.
  • Stock vocabulary: delve, tapestry, testament, realm, navigate (figurative), leverage, seamless, robust, pivotal, crucial, elevate, unlock, embark, foster, "plays a vital role". Use the plain word or a concrete claim.
  • Reflexive triads: "fast, reliable, and scalable" in every paragraph. Use one, two or four items when that is the true number.
  • Em-dash habit: one or two per page is fine; one per sentence is a signature. Use commas, periods, parentheses or colons.
  • Hedging and qualifiers: "it's important to note", "arguably", "can potentially", "it may be worth considering". Commit, or state the real uncertainty once.
  • Symmetrical structures: "It's not just X, it's Y." "Whether you're A or B..." "From X to Y". Fine once; not as a template.
  • Generic conclusions: "In conclusion", "Ultimately", "By following these tips, you can...". End on the last useful point, a concrete next step, or a line with a view.
  • Over-formatting: headings on a 200-word email, bold on every other phrase, bullets for things that read better as a sentence, emoji as bullet markers.
  • Empty intensifiers and praise: "truly", "incredibly", "game-changer", "revolutionary", "I hope this email finds you well".
  • Uniform rhythm: every sentence 15-20 words, every paragraph three sentences.

What to do instead

  • Be specific. Replace abstractions with a number, a name, a place, an example, a short anecdote. "Saves time" -> "cut our weekly report from 3 hours to 40 minutes".
  • Vary rhythm. Mix short sentences with longer ones. A fragment now and then. Let a paragraph be one line when it lands a point.
  • Active voice, real subjects. "We shipped it late" not "Delays were experienced".
  • Say it once. Cut the sentence that restates the previous one.
  • Use the words the reader uses. Plain verbs, everyday nouns, the jargon of the field only when the reader shares it.
  • Allow opinion and texture where the author would: a preference, a doubt, a small aside. Humans are not perfectly balanced.
  • Format for the medium. Chat and email: plain paragraphs. Docs: headings if they help navigation. Social: see the platform skills.
  • Idiom belongs to the language. Do not translate English idioms or English punctuation habits into other languages; write natively.

Process

  1. Read the whole text first. Note the core message in one sentence.
  2. Cut: openers, closers that summarize, duplicate sentences, hedges.
  3. Replace: stock words and vague claims with plain words and specifics. If the specific fact is unknown, ask or leave a clear placeholder like [number]; never invent facts, quotes or anecdotes.
  4. Restructure: break the rhythm, remove unneeded lists and headings.
  5. Read it aloud (mentally). Anything you would not say to a colleague, rewrite.
  6. Check length: usually shorter than the original.

Before/after examples: references/examples.md.

Output

  • Return the rewritten text only, unless the user asked for explanations. If they want to learn, add a short list of the main changes afterwards.
  • Keep the original language, meaning, facts and any required terms, names or links.
  • Do not claim the text will pass AI detectors; detectors are unreliable. The goal is good writing.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.