humanizer

Rewrites text to remove AI-generated writing patterns and restore natural voice.

10|2|Updated Jan 24, 2026
One-click install
npx skills add https://github.com/nrdxp/predicate --skill humanizer-nrdxp
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: humanizer
Source: https://github.com/nrdxp/predicate/tree/main/skills/humanizer
Command: npx skills add https://github.com/nrdxp/predicate --skill humanizer-nrdxp

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? AI-generated text carries recognizable patterns—inflated significance claims, promotional language, vague attributions, and formulaic structures—that make writing sound robotic and erode reader trust. This Skill identifies and rewrites those patterns so documentation, comments, and summaries read as natural human prose. ## Core Features & Use Cases - Pattern Detection: Scans text for 24 documented AI writing patterns, including overused vocabulary ("delve", "tapestry", "pivotal"), superficial -ing phrases, rule-of-three lists, em dash overuse, and sycophantic tone. - Voice Restoration: Goes beyond removing bad patterns by injecting personality—varied sentence rhythm, first-person perspective where appropriate, and specific concrete details instead of vague claims. - Before/After Rewriting: Provides rewritten text plus an optional summary of changes, grounded in Wikipedia's "Signs of AI writing" guide maintained by WikiProject AI Cleanup. - Use Case: You drafted a README with an AI assistant and it reads like a press release full of "nestled in the heart of" and "stands as a testament to". Run the humanizer to strip the slop and get prose that sounds like a person wrote it. ## Quick Start Humanize this project summary and remove any AI writing patterns while keeping the technical details intact.

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?

Scan the text for known AI patterns like overused vocabulary ("delve", "pivotal", "landscape"), superficial -ing phrases, and rule-of-three lists, then rewrite each instance with specific concrete details. Vary sentence length and add genuine perspective rather than neutral reporting.

What are the most common signs of AI writing?

The most frequent tells include inflated significance language ("stands as a testament"), promotional adjectives ("vibrant", "nestled"), vague attributions ("experts believe"), em dash overuse, boldface-heavy lists, and generic upbeat conclusions. This Skill covers 24 documented patterns.

Does humanizing text change its meaning?

No. The rewriting process preserves the core message and intended tone while replacing AI-isms with natural alternatives. It favors specific facts over vague claims, so meaning typically becomes clearer rather than distorted.

When should I not remove em dashes and boldface from text?

The em dash and boldface rules apply only to reader-facing prose like READMEs, guides, and tutorials. Normative files such as rules, specs, and skill definitions use these conventions by design and are explicitly out of scope.

Why does AI-cleaned text still sound robotic sometimes?

Removing bad patterns is only half the job; sterile, voiceless writing is equally obvious. The Skill also restores soul by varying sentence rhythm, acknowledging mixed feelings, using first person where appropriate, and adding specific emotional detail.