stet

Remove 27 AI-writing patterns from text to restore authentic voice.

Updated Feb 19, 2026
One-click install
npx skills add https://github.com/Cronan/claude-code-prompts --skill stet
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: stet
Source: https://github.com/Cronan/claude-code-prompts/tree/main/skills/stet
Command: npx skills add https://github.com/Cronan/claude-code-prompts --skill stet

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Stet identifies and removes a suite of AI-writing patterns that make text sound generic, helping authors recapture a distinctive voice.

Core Features & Use Cases

  • Detects and strip 27 AI-writing patterns across structure, vocabulary, and polish to preserve meaning while improving voice.
  • Suitable for articles, reports, and drafts that were AI-assisted, enabling editors to reintroduce personality, nuance, and author-specific cadence.
  • Use Case: An editor pastes AI-generated copy and uses Stet to produce text that sounds like the original author.

Quick Start

Copy the stet skill into your repository and run /stet on your document.

Frequently Asked Questions about stet

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

FAQPage Schema
How do I remove AI writing patterns from my generated drafts?

To remove AI writing patterns from generated drafts, you can apply a prose-editing skill that detects and strips 27 specific structural, vocabulary, and polish patterns to restore authentic human voice while preserving meaning.

What is prose editing for AI-assisted text?

Prose editing for AI-assisted text is the process of identifying and removing generic AI-writing patterns to reintroduce personality, nuance, and author-specific cadence to articles and reports.

How do I make AI-generated copy sound like the original author?

You make AI-generated copy sound like the original author by processing the text through a pattern-stripping tool that enforces varied rhythm and authentic authorial tone without overcorrecting the meaning.

Can I use automated editing on AI-assisted reports without losing meaning?

Yes, you can use automated editing on AI-assisted reports without losing meaning, as the process preserves the original intent while removing 27 defined AI-writing patterns and avoiding overcorrection.

What are the limitations of automated AI pattern removal?

The main limitation of automated AI pattern removal is the risk of overcorrection, which is why the process specifically enforces a defined set of 27 patterns and preserves meaning to maintain text integrity.