humanizer

Detect and rewrite AI writing patterns in text to sound natural and human.

283|23|Updated Jul 23, 2026
One-click install
npx skills add https://github.com/NulightJens/humanizer-stack --skill humanizer-nulightjens
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: humanizer
Source: https://github.com/NulightJens/humanizer-stack/tree/main/skills/humanizer
Command: npx skills add https://github.com/NulightJens/humanizer-stack --skill humanizer-nulightjens

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? AI-generated text carries recognizable patterns—inflated symbolism, promotional language, em dash overuse, rule of three, vague attributions—that make writing read as machine-made. This Skill identifies those surface-level tells and rewrites them into natural, human-sounding prose while preserving meaning and tone. ## Core Features & Use Cases - Pattern Detection and Rewriting: Identifies 24 documented AI writing patterns from Wikipedia's "Signs of AI writing" guide, including AI vocabulary words, negative parallelisms, superficial -ing analyses, and filler phrases, then rewrites each with natural alternatives. - Copy-Specific Tells: A supplementary reference covers public-facing copy (landing pages, marketing sites, social posts) with tells ranked by a 3.2M-post Reddit analysis, including the em dash, hype vocabulary, and sycophantic openers. - Voice Injection: Goes beyond removing bad patterns by adding opinions, varied sentence rhythm, and first-person perspective so the result has personality rather than sterile neutrality. - Use Case: Paste an AI-drafted blog post or landing page and receive a rewritten version with specific changes summarized, ready for publication. ## Quick Start Humanize this AI-generated draft and list the patterns you fixed.

Frequently Asked Questions about humanizer

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I remove AI writing patterns from text?

Provide the text and the Skill scans it for 24 documented patterns like inflated symbolism, AI vocabulary, em dash overuse, and negative parallelisms, then rewrites each problem section. It returns the humanized version plus an optional summary of changes made.

What AI writing tells does this detect?

It detects content patterns like undue emphasis on significance and vague attributions, language patterns like copula avoidance and synonym cycling, and style patterns like em dash overuse, boldface abuse, curly quotes, and title case headings, based on Wikipedia's Signs of AI writing guide.

Does humanizing text guarantee it passes AI detectors?

No. This is the surface pass covering words and phrasing only. The StoryScope study showed professional surface rewriting moved AI detection by only 1.6 points, while structural features detect AI text at 93.2% F1, so a structural pass is needed afterward.

Can I use this for marketing copy and landing pages?

Yes. A dedicated copy-tells reference covers public-facing copy with tells ranked by a 3.2M-post Reddit analysis, including the em dash, hype vocabulary like "supercharge" and "unlock", and sycophantic openers like "Great question!".

What is the difference between humanizer and structural-humanizer?

Humanizer fixes surface-level words, phrasing, and punctuation, while structural-humanizer fixes discourse-level tells like stated lessons, tidy narrative arcs, and embodied-emotion performance. Run humanizer first, then structural-humanizer for text that must read as human.