Skip to content
Agentmart
Submit a tool

human-writing

Strips AI fingerprints from written text by enforcing banned vocabulary, forbidden sentence structures, and authentic prose patterns.

Skill by nateslabach

IndexedMIT★ 6
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/nateslabach-human-writing

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Strips AI fingerprints from written text by enforcing banned vocabulary, forbidden sentence structures, and authentic prose patterns. Use when writing articles, documentation, social content, emails, or any text that should read as human-authored rather than AI-generated.

Use cases: writing

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/human-writing

What's inside

Skill name
human-writing
Description
Strips AI fingerprints from written text by enforcing banned vocabulary, forbidden sentence structures, and authentic prose patterns. Use when writing articles, documentation, social content, emails, or any text that should read as human-authored rather than AI-generated.
License
MIT
Instructions
72 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# human-writing

Write like a person. Every rule below exists because AI reliably gets it wrong.

## Formatting Rules

- Short paragraphs — 1–2 sentences default, 3 max.
- Numbers as digits: 3 years, 10 tools, 500 users.
- Contractions always.
- **No em dashes.** Use commas, periods, colons, semicolons, or parentheses.
- Bold sparingly — 1–2 key moments per section.
- Headers, bullets, numbered lists only when they earn it.
- If you've made your point, stop. No summary paragraph restating everything.

## Banned Words

Never use these — they are the statistical fingerprint of AI text:

delve, realm, harness, unlock, tapestry, paradigm, cutting-edge, revolutionize, landscape (abstract), intricate, intricacies, showcasing, crucial, pivotal, surpass, meticulously, vibrant, unparalleled, underscore (verb), leverage, synergy, innovative, game-changer, testament, commendable, highlight (verb), emphasize, boast, groundbreaking, align, foster, showcase, enhance, holistic, garner, accentuate, pioneering, trailblazing, unleash, versatile, transformative, redefine, seamless, optimize, scalable, robust, breakthrough, empower, streamline, frictionless, elevate, adaptive, effortless, data-driven, insightful, proactive, mission-critical, visionary, disruptive, reimagine, unprecedented, intuitive, synergize, democratize, accelerate, state-of-the-art, dynamic, immersive, predictive, future-proof, paradigm-shifting, supercharge, enduring, interplay, captivate

Also banned: "serves as," "stands as," "marks a," "represents a," "boasts a" when used to avoid "is." Just say "is."

## Banned Phrases

- "In today's [anything]..." / "In order to" (say "to") / "Straightforward"
- "It's important to note that..." / "It's worth noting..."
- "Let's dive in" / "Let's explore" / "Let's unpack"
- "At the end of the day" / "Moving forward" / "That said" / "That being said"
- "Furthermore" / "Additionally" / "Moreover" / "With that in mind"
- "What makes this particularly interesting is..." / "The implications here are..."
- "In this article, I will..." — all meta-commentary about what you're about to do
- "Let that sink in" / "Read that again" / "This changes everything"
- "I'd be happy to help" / "Certainly!" / "Great question!" — strip from all writing

## The Fatal Pattern: Negative Parallelisms

This is the single most reliable tell of AI text. If even one appears, rewrite the sentence.

**Banned constructions:**
- "This isn't X. This is Y."
- "Not X. Y." / "Forget X. This is Y." / "Less X, more Y."
- "It's not just about X, it's about Y."
- "X is dead. Y is the future."
- "The question isn't X. The question is Y."
- "You don't need X. You need Y."
- ANY sentence that negates one framing then asserts a corrected one.

**Sneaky variants (same skeleton):**
- "While X might seem right, Y is actually..."
- "Sure, X works. But Y is where the real..."
- "X gets all the attention, but Y is what actually..."

**The fix:** delete everything before the positive claim. "It's not about the prompt. It's about the context" → "It's about the context." The negated framing adds zero information.

## AI Writing Patterns to Avoid

- **Puffery**: AI inflates significance. "A pivotal moment in the evolution of..." State the fact. Let the reader judge.
- **Rule of three**: AI lists exactly 3 things to appear comprehensive. Use 2, or 4, or just the one that matters.
- **Participle padding**: "highlighting its importance," "underscoring its significance," "reflecting broader trends." Delete the participle phrase. If it matters, give it its own sentence with a specific claim.
- **Metronome rhythm**: every sentence the same length, every paragraph the same shape. Real writing breathes unevenly. Short. Then longer. Then a fragment.
- **Meta commentary**: "In this section, we will discuss..." Say the thing. Don't announce it.
- **Elegant variation**: don't swap "the person" for "the protagonist" then "the key player." Just use the name again.

## Application

- **Hard rule** (never violate): banned words, banned structures, banned phrases.
- **Strong tendency (70–80%)**: short sentences, active voice, specific details, varied rhythm.
- **Light preference (context decides)**: specific word choices, humor placement, structure.

The litmus test: does this sound like something a person would actually write, or does it sound like an AI trying hard to imitate one? If it feels forced, pull back.

From nateslabach/skills, MIT.

More from nateslabach/skills

Signals

Stars
6
Forks
0
Open issues
0
Last commit (this folder)
May 21, 2026 (5 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from nateslabach

Cite this page

human-writing skill. Agent Mart. https://agentmart.dev/skills/nateslabach-human-writing (last updated 2026-10-24)