humanizer

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

Updated Aug 1, 2026
One-click install
npx skills add https://github.com/nicogoldberg1-rgb/opensourcing-release --skill humanizer-nicogoldberg1-rgb
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: humanizer
Source: https://github.com/nicogoldberg1-rgb/opensourcing-release/tree/main/skills/humanizer
Command: npx skills add https://github.com/nicogoldberg1-rgb/opensourcing-release --skill humanizer-nicogoldberg1-rgb

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? AI-generated text carries recognizable tells—inflated significance, promotional language, em dash overuse, rule-of-three structures, and filler phrases—that make prose sound machine-written. This Skill identifies 33 documented patterns from Wikipedia's "Signs of AI writing" guide and rewrites text to read naturally while preserving every fact from the source. ## Core Features & Use Cases - Pattern Detection and Removal: Scans for 33 numbered AI writing patterns across content, language, style, and communication categories, then rewrites with a draft → audit → final loop. - Voice Calibration: When given a writing sample, matches the author's sentence lengths, vocabulary, and punctuation habits instead of scrubbing to a generic style. - No-Fabrication Guarantee: The rewrite never introduces facts, names, numbers, dates, or citations absent from the source text. - Use Case: After drafting cold outreach emails with an AI assistant, run the text through this Skill to strip chatbot artifacts like "I hope this helps" and promotional phrasing before sending to real recipients. ## Quick Start Paste your AI-drafted text and ask Claude to humanize it so it reads like natural human writing.

Frequently Asked Questions about humanizer

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

FAQPage Schema
How do I make AI-generated text sound more human?

Paste the text and ask for it to be humanized. The skill scans for 33 known AI writing patterns, produces a draft rewrite, audits it for remaining tells and fabricated facts, then delivers a final version that preserves all source information.

What AI writing patterns does this detect?

It detects 33 patterns from Wikipedia's Signs of AI writing guide, including inflated significance, promotional language, vague attributions, em dash overuse, rule of three, AI vocabulary words, negative parallelisms, filler phrases, and chatbot artifacts like "I hope this helps."

Does humanizing text require an API key or external service?

No. The skill is a Markdown prompt only, with no scripts, API keys, or MCP servers. It runs entirely within the AI assistant's context using the pattern guide embedded in its instructions.

Can it match my personal writing style instead of a generic tone?

Yes. Provide a sample of your own writing and the skill analyzes its sentence lengths, vocabulary, punctuation, and recurring phrases, then matches those habits. A supplied sample overrides the default style rules, including the em dash ban.

Will rewriting change the facts in my text?

No. A hard rule forbids introducing any fact, name, number, date, quote, or citation not present in the source. The audit step explicitly checks the rewrite for fabricated details before delivering the final version.