Systems Lab

Agent skill

humanizer

Humanize AI-sounding writing and remove common AI-generated writing tells while preserving meaning, audience, and brand voice.

slowingSelf-containedActs undeclared733 words

Filed under Content and SEO.

From manojbajaj95/claude-gtm-plugin · 51 skills · 92 · pushed 2026-05-18

What it does when it runs

Humanize AI-sounding writing and remove common AI-generated writing tells while preserving meaning, audience, and brand voice. Use when editing text to sound more natural, less generic, less promotional, less formulaic, or closer to the user's own writing. Triggers: humanize this, make this less AI, make this sound human, remove AI writing patterns, rewrite in my voice, make this more natural, de-AI this, make this less ChatGPT, fix robotic writing, improve authenticity, reduce generic AI tone.

Read from the skill and the 0 files bundled beside it. A skill’s own description is written to be selected by an agent, so it describes the job and not the dependencies.

Keys and connectors you must supply
None found.
Hosts it reaches
No third-party host appears in the skill or its bundled files.
Tool permissions it declares
No allowed-tools in the frontmatter. It does act, so it runs under whatever permissions your session already grants.
Actions present in the files
writes files

Ask about humanizer

Opens your assistant with this page's verified links already in the prompt.

Is this safe to install?ClaudeChatGPT
Adapt it to my stackClaudeChatGPT
What else do I need for it to workClaudeChatGPT
Rather ask a human? Talk to Cheetah
git clone --depth 1 --filter=blob:none --sparse https://github.com/manojbajaj95/claude-gtm-plugin.git /tmp/claude-gtm-plugin
git -C /tmp/claude-gtm-plugin sparse-checkout set "skills/humanizer"
mkdir -p ~/.claude/skills/humanizer
cp -R "/tmp/claude-gtm-plugin/skills/humanizer/." ~/.claude/skills/humanizer/

Picked up without a restart. A project skill of the same name is shadowed by your personal one. For one repository only, swap ~/.claude/skills for .claude/skills. Claude Code docs ↗

Or take the whole library

This repo ships a .claude-plugin manifest, so Claude Code can install all 51 skills at once. Plugin skills are invoked as /<plugin>:<skill>, so they never collide with your own.

/plugin marketplace add manojbajaj95/claude-gtm-plugin
/plugin

The folder is the same in every client that implements the format — 46 of them — so if yours is not above, only the destination changes.

Reproduced in full from manojbajaj95/claude-gtm-plugin/blob/3c308fa4f7879d13d3d3df394247eabf6d26b684/skills/humanizer/SKILL.md, which is licensed MIT (repository). 733 words, 18 headings.

Humanizer

Workspace Context

Read bootstrap context before asking questions: strategy/brand.md for brand, audience, offer, channels, tools, constraints, and metrics; about/me.md for personal voice; content/ideas.md and content/calendar.md for content planning. Use legacy product-marketing context files only as fallback. Save generated drafts to content/<platform>/drafts/YYYY-MM-DD_short-topic-slug.md, and route durable learnings back to strategy/brand.md, about/me.md, or content/ideas.md.

Operating Contract

This skill is self-contained for its frontmatter scope: use its local instructions, references, scripts, and assets as the playbook; ask only for missing task-specific inputs; hand off to adjacent skills instead of expanding scope; and return an actionable artifact, decision, plan, draft, or diagnostic.

You are a careful editor. Your job is to make writing sound like a real person wrote it, not to make it more polished for its own sake.

Inputs

Use the text the user provides. Before asking questions, check:

  • about/me.md for personal writing style
  • strategy/brand.md for brand voice, audience, and claims
  • Any user-provided sample, file, or prior draft

Ask only if the target voice, audience, or destination is genuinely unclear.

Workflow

  1. Read for intent: identify the audience, job of the piece, desired action, and non-negotiable claims.
  2. Find AI tells: mark the patterns that make the writing feel synthetic.
  3. Rewrite once: preserve meaning, claims, and structure where useful, but remove formulaic language.
  4. Run an anti-AI audit: ask, "What still makes this obviously AI generated?" Note the remaining tells briefly.
  5. Rewrite again: produce the final version with the audit fixed.

Voice Calibration

If about/me.md or a user-provided sample exists, match it:

  • sentence length and rhythm
  • paragraph openings
  • punctuation habits
  • directness, humor, edge, warmth, or restraint
  • preferred plain words over polished substitutes
  • transitions, asides, and recurring phrasing

Do not "upgrade" casual wording into corporate language. If the user writes "stuff," do not automatically change it to "initiatives" or "components."

Patterns to Remove

Inflated significance

Watch for: "serves as," "stands as," "testament," "pivotal," "underscores," "broader trend," "evolving landscape," "lasting impact."

Fix by stating the concrete fact and why it matters, without pretending every detail is historic.

Promotional filler

Watch for: "vibrant," "rich," "groundbreaking," "seamless," "powerful," "must-have," "nestled," "in the heart of," "showcasing," "commitment to."

Fix with specific evidence, plain nouns, and direct claims.

Fake depth

Watch for -ing add-ons: "highlighting," "underscoring," "reflecting," "encompassing," "fostering," "ensuring."

Fix by splitting the sentence or deleting the fake analysis.

Vague attribution

Watch for: "experts say," "industry reports suggest," "observers note," "some critics argue."

Fix with a named source, a real example, or remove the claim.

AI vocabulary clusters

Watch for repeated use of: "additionally," "delve," "crucial," "intricate," "interplay," "landscape," "robust," "valuable," "key," "enhance," "leverage."

Fix with shorter, more ordinary words.

Formulaic structure

Watch for:

  • forced groups of three
  • "not only X but Y"
  • "it is not just X, it is Y"
  • false "from X to Y" ranges
  • generic "challenges and future prospects" sections
  • upbeat closing filler

Fix by keeping only the real argument.

Chatbot residue

Remove conversational artifacts that belong to a chat response, not final copy:

  • "Great question"
  • "I hope this helps"
  • "let me know if..."
  • "here is an overview"
  • "let's dive in"
  • knowledge-cutoff disclaimers

Over-styled formatting

Reduce:

  • excessive boldface
  • emoji decorations
  • title-case headings when sentence case fits
  • inline-header bullet lists
  • em dash overuse
  • curly quotes if the project uses straight quotes

Add Humanity

Removing AI tells is not enough. Add signs of a real writer where appropriate:

  • a specific opinion
  • a concrete example
  • a sentence with uneven rhythm
  • a useful caveat
  • a direct first-person line when the format allows it
  • a small imperfection that sounds natural, not sloppy

Do not invent personal experiences, customers, metrics, quotes, or sources.

Output Format

Return:

  1. Draft rewrite
  2. Remaining AI tells: brief bullets from the anti-AI audit
  3. Final rewrite
  4. What changed: optional, only if useful

If editing a file, save the final version to the correct destination and summarize the changes. For new drafts, use content/<platform>/drafts/YYYY-MM-DD_short-topic-slug.md unless the user asks for inline output only.

Related Skills

  • copywriting-core for writing or rewriting conversion copy from scratch
  • blog-writing-specialist for long-form blog drafts
  • linkedin for LinkedIn-specific posts and profile content
  • content-creation-and-marketing for cross-channel draft production

Other skills for the same job

Different authors, same problem. Matched on the words in the skill name, across every library in the catalogue except this one.

Need help setting it up?

This page tells you what humanizer does and what it needs. Cheetah builds the agent setup it runs inside: data, CRM, sequencing and the guardrails.

Book a call →

The directory stays free. There is nothing gated behind this.