write-for-humans

Detect and rewrite AI-slop prose while preserving the author's voice.

3|Updated Jan 14, 2014
One-click install
npx skills add https://github.com/prateek/dotfiles --skill write-for-humans
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: write-for-humans
Source: https://github.com/prateek/dotfiles/tree/main/.agents/skills/write-for-humans
Command: npx skills add https://github.com/prateek/dotfiles --skill write-for-humans

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

AI-generated prose often sounds generic and stilted. This Skill identifies AI-slop patterns and rewrites text to read as human-written, preserving meaning and voice.

Core Features & Use Cases

  • Detects AI-slop patterns and performs rule-based rewrites to remove scaffolding, inflated vocabulary, and negation traps.
  • Preserves authorial voice and technical meaning while improving readability across docs, READMEs, emails, PR descriptions, and chat replies.
  • Use cases: rewriting drafts, reviewing long replies, polishing documentation, and cleaning up code comments.

Quick Start

Paste prose and instruct me to rewrite it so it reads as human, not AI-generated.

Frequently Asked Questions about write-for-humans

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

FAQPage Schema
How do I make AI-generated prose read like human writing?

To make AI-generated prose read like human writing, the system detects AI-slop patterns and applies a seven-rule rewrite process to remove scaffolding and inflated vocabulary. It preserves the author's voice while maintaining the original technical meaning.

What are common AI-slop patterns in documentation and emails?

Common AI-slop patterns in documentation and emails include generic cadence, inflated descriptors, and unnecessary scaffolding. The system identifies these patterns and rewrites the text by applying rule-based corrections to improve readability.

How do I rewrite a README to remove AI cadence without losing technical meaning?

To rewrite a README and remove AI cadence without losing technical meaning, the system strictly preserves the original intent and technical content. It applies a seven-rule process to strip AI-like phrasing while keeping the authorial voice intact.

Can I preserve my authorial voice when humanizing AI drafts?

Yes, you can preserve your authorial voice when humanizing AI drafts. The rewriting process strictly maintains the author's original voice and technical content, ensuring the rewritten text reflects the intended tone rather than a generic style.

Does this approach work for cleaning up long chat replies and PR descriptions?

Yes, this approach works for cleaning up long chat replies and PR descriptions. The system is designed to process docs, READMEs, emails, PRs, and long user replies, ensuring the rewritten text maintains original intent without AI-like cadence.

When should I not use a rule-based prose rewrite?

You should not use a rule-based prose rewrite when the primary goal is to drastically alter the original intent or technical meaning. The process is designed to preserve meaning and voice, so it may not suit text requiring structural transformation rather than cadence correction.