agentleFS
Sign inStart free
spencerpaulyspencerpauly/skills-repo/skills/write-like-a-human/SKILL.md

write-like-a-human

Skill25 starsChanged 7 months agoMITView on GitHub

Strip the AI-isms out of any text — em-dash overuse, "It's not just X, it's Y", inflated significance, sycophantic openers, rule-of-three padding. Use when the user says "humanize this", "make this less AI", "tighten this", or pastes obviously-AI prose and asks for a rewrite.

4 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions4 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 4

    1. thebpandey/write-like-a-human★ 9same name
    2. seangeng/skills★ 7same name
    3. jazzsequence/claude-write-like-a-human-skill★ 2same name
    4. brandhaug/skills★ 0same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository5 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

3 rules. 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 4 edited versions we read keep these.

  • 3

    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:

    Don't replace AI mush with new AI mush.

    line 48
  2. Distinctive:

    Don't sand off the user's voice.

    line 49
  3. Distinctive:

    Don't add em-dashes back in.

    line 50

“Common” is measured: kept by at least 6 in 10 of the 4 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

Write Like a Human

Rewrite text so it stops sounding like a language model wrote it. Wraps the open-source humanizer skill, which catalogs ~30 specific tells (em-dash overuse, "stands as a testament", -ing analyses, rule-of-three padding, inline-bold lists, and so on).

When to use

  • The user pastes a chunk of writing and asks to "humanize", "de-AI", "tighten", or "make this sound like me".
  • The user asks for editing on something they wrote with an LLM.
  • A blog post / launch email / README draft has tells the user wants gone.

Steps

  1. Load the reference list of patterns the first time this skill activates in a session. On Spencer's machine, read the canonical local copy (newer than upstream):
    cat ~/.agents/skills/humanizer/SKILL.md
    Otherwise pull it from upstream:
    curl -fsSL https://raw.githubusercontent.com/blader/humanizer/main/SKILL.md
    There are ~30 patterns; you don't need to memorize them, just keep them visible while editing.
  2. Ask for a voice sample if you don't already have one. One paragraph the user wrote themselves is enough. Match cadence, vocabulary, and how they handle transitions. For content written as Spencer, the voice layer is the spencers-writing-style skill in the stack repo (.agents/skills/spencers-writing-style/) — use it instead of asking.
  3. Do a first pass removing the highest-signal tells in this order:
    1. Sycophantic openers ("Great question!", "Absolutely!").
    2. Em-dashes that should be commas, periods, or parens.
    3. "It's not just X, it's Y" / "Not only … but …" parallelisms.
    4. Inflated significance words: testament, pivotal, landscape, vibrant, tapestry, underscore.
    5. Rule-of-three lists that pad ideas into groups of three.
    6. Inline-bold vertical lists (- **Speed:** …).
  4. Do a second "audit" pass. Re-read your draft and ask out loud: "What still makes this sound AI-generated?" Answer briefly, then revise once more. This step is the difference between "cleaned up" and "actually sounds human".
  5. Return three things:
    • A draft rewrite.
    • A short list of remaining tells you noticed during the audit.
    • A final rewrite with those fixed.

Don't

  • Don't replace AI mush with new AI mush. If the original said "stands as a testament to" and you change it to "represents a key milestone in", you haven't fixed anything.
  • Don't sand off the user's voice. Quirks, opinions, and the occasional sentence fragment are the point.
  • Don't add em-dashes back in. If you must use a punctuation break, prefer a period.

Reference

Patterns and examples: https://github.com/blader/humanizer

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.

Check your setup, free