avoid-ai-writing

Detects and rewrites AI writing patterns using 21 pattern categories and a 43-entry replacement table.

2|1|Updated Aug 19, 2026
One-click install
npx skills add https://github.com/lanceyuu/mimiwork --skill avoid-ai-writing-lanceyuu
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: avoid-ai-writing
Source: https://github.com/lanceyuu/mimiwork/tree/main/coworker/skills/builtin/avoid-ai-writing
Command: npx skills add https://github.com/lanceyuu/mimiwork --skill avoid-ai-writing-lanceyuu

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? AI-generated drafts often contain recognizable patterns—hedging phrases, hollow intensifiers, template transitions, and formulaic structures—that make text sound machine-generated. This Skill audits prose for these tells and rewrites it into plainer, more natural language. ## Core Features & Use Cases - Pattern Detection: Flags 21 categories of AI-isms including em dash overuse, bold overuse, hedging, rule of three, significance inflation, vague attributions, and chatbot artifacts. - Replacement Table: Maps 43 flagged words and phrases to plainer alternatives, such as leverage→use, utilize→use, and robust→reliable. - Four-Section Output: Returns issues found, a rewritten version, a summary of changes, and a second-pass audit to catch surviving tells. - Use Case: Before publishing a blog post drafted with AI, run an audit to strip phrases like "In today's rapidly evolving landscape" and "embarking on a pivotal journey" and replace them with direct, human-sounding prose. ## Quick Start Audit this draft for AI writing patterns and rewrite it to sound natural.

Frequently Asked Questions about avoid-ai-writing

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

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

Run an audit that flags AI-isms across 21 pattern categories, then apply the rewrite that replaces flagged terms with plainer alternatives. The output includes the issues found, the rewritten text, a change summary, and a second-pass audit.

What words and phrases does AI writing detection flag?▼

The replacement table covers 43 entries such as leverage→use, utilize→use, and robust→reliable. It also flags template phrases, hollow intensifiers, significance inflation, and transitions like "Moreover" that commonly appear in generated text.

Can this detect AI-generated code?▼

No, detection applies only to prose, not code. The pattern categories target natural-language writing tells such as hedging, formulaic structures, and promotional language.

Are all flagged words always wrong to use?▼

No, pattern matching is guideline-based rather than absolute. Some flagged words are acceptable in context, and the replacement table offers suggestions where the best choice depends on the surrounding sentence.

What are the limitations of AI writing detection?▼

It cannot verify factual claims or find real citations to replace vague attributions. It also does not detect AI-generated code, and flagged patterns require human judgment about context.