agentleFS
Sign inStart free
styfinitystyfinity/linkedin-engine/skills/linkedin-humanizer/SKILL.md

linkedin-humanizer

Skill9 starsChanged 4 months agoMITView on GitHub

Strip the AI fingerprints out of any draft (a post, a DM, a comment) so it reads like a real person wrote it. Use before publishing anything written or edited with AI.

7 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

7 rules. 0 are common, 7 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 7 edited versions we read keep these.

  • 7

    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:

    Remove every em-dash. Replace with a comma, a period, or " - ".

    line 16
  2. Distinctive:

    Delete fake-candid openers: "Let me be honest", "Here's the thing", "I'll be the first to admit".

    line 19
  3. Distinctive:

    Put the human back: one casual run-on with a comma splice, plain low-reading-level words, one direct un-hedged opinion, and a flat …

    line 21
  4. Distinctive:

    Keep "I" not "we" in the hook.

    line 22
  5. Distinctive:

    Never change the meaning or the claims.

    line 28
  6. Distinctive:

    If anything still reads like AI after the pass, flag the specific phrase, do not silently let it through.

    line 30
  7. Distinctive:

    No em-dashes in the output.

    line 31

“Common” is measured: kept by at least 6 in 10 of the 7 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

LinkedIn Humanizer

The draft that reads like a machine wrote it gets ignored. This skill removes the tells and puts the human voice back in, without touching the meaning.

Inputs

  • Any draft (a post, a DM, a comment): $ARGUMENTS
  • The brief (persona voice, pains, offer) loads automatically, so use it to match the operator's real register.

Do this

  1. Remove every em-dash. Replace with a comma, a period, or " - ".
  2. Strip banned AI vocabulary: leverage, delve, landscape, navigate, utilize, foster, robust, supercharge, seamless, unlock, elevate, harness, empower, streamline, "game-changer", and any thesaurus word a person would not say out loud.
  3. Break up rule-of-three symmetric lists. Real people do not write in perfect triplets.
  4. Delete fake-candid openers: "Let me be honest", "Here's the thing", "I'll be the first to admit". Cut straight to the point.
  5. Vary the sentence length. Uniform rhythm is the loudest AI tell. Put a short one next to a long one.
  6. Put the human back: one casual run-on with a comma splice, plain low-reading-level words, one direct un-hedged opinion, and a flat statement (not a question) to end on.
  7. Keep "I" not "we" in the hook. The operator is one person, not a brand.

Output

The cleaned draft first. Then a short bullet diff: what was changed and why. Then a one-line verdict: PUBLISH, or STILL READS AI - fix these (name the exact phrases).

Rules

  • Never change the meaning or the claims. Keep all numbers exactly as written.
  • This is a polish pass, not a rewrite. If a line is already human, leave it.
  • If anything still reads like AI after the pass, flag the specific phrase, do not silently let it through.
  • No em-dashes in the output. Obviously.
  • This skill only cleans drafts. It does not publish. The operator approves and posts.

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