writing-anti-ai

Detect and rewrite AI writing patterns in English and Chinese prose.

1|Updated Mar 5, 2026
One-click install
npx skills add https://github.com/Clay-HHK/claude-skills --skill writing-anti-ai-clay-hhk
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: writing-anti-ai
Source: https://github.com/Clay-HHK/claude-skills/tree/main/writing-anti-ai
Command: npx skills add https://github.com/Clay-HHK/claude-skills --skill writing-anti-ai-clay-hhk

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? AI-generated text contains predictable patterns—inflated symbolism, promotional language, vague attributions, and formulaic structures—that make prose sound robotic and generic. This Skill identifies and removes these patterns while injecting authentic human voice. ## Core Features & Use Cases - Pattern Detection: Identifies 20+ AI writing patterns based on Wikipedia's "Signs of AI writing" guide, including AI vocabulary, negative parallelisms, rule of three, and superficial -ing analyses. - Bilingual Support: Handles both English and Chinese text with language-specific pattern references and examples. - Scoring System: Rates text on a 50-point scale across directness, rhythm, trust, authenticity, and density. - Use Case: Paste an AI-drafted product announcement and receive a rewritten version with specific details, varied sentence rhythm, and natural voice instead of "serves as a testament" and "pivotal role in the evolving landscape." ## Quick Start Humanize this text and remove all AI writing patterns from the attached draft.

Frequently Asked Questions about writing-anti-ai

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

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

Scan the text for common AI patterns like inflated symbolism, promotional language, and vague attributions, then rewrite each flagged section with direct statements and specific details. This Skill provides checklists and before/after examples for both English and Chinese text.

What are the most common signs of AI-generated writing?▼

Common signs include phrases like "serves as a testament," "pivotal role," and "evolving landscape," plus negative parallelisms ("It's not just X, it's Y"), rule-of-three structures, em dash overuse, and vague attributions like "experts believe." The full list is based on Wikipedia's Signs of AI writing guide.

Does this AI text humanizer support Chinese writing?▼

Yes, it supports both English and Chinese with dedicated pattern references for each language. Chinese-specific patterns include phrases like "作为……的证明", "值得注意的是", and "此外", with localized before/after examples.

How is the AI writing detection scored?▼

Text is rated 1-10 on five dimensions: directness, rhythm, trust, authenticity, and density, for a total of 50 points. Scores of 45-50 indicate AI patterns are removed, while below 35 means the text needs revision.

Can removing AI patterns make writing sound too plain?▼

Yes, which is why the Skill pairs pattern removal with voice injection: adding opinions, varied sentence rhythm, first-person perspective where appropriate, and acknowledgment of mixed feelings. Sterile, voiceless writing is treated as a failure mode, not a goal.