stop-slop

Detects and removes AI writing patterns from prose while preserving author voice.

5|Updated Sep 24, 2017
One-click install
npx skills add https://github.com/madyankin/dotfiles --skill stop-slop-madyankin
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: stop-slop
Source: https://github.com/madyankin/dotfiles/tree/main/.config/agents/skills/stop-slop
Command: npx skills add https://github.com/madyankin/dotfiles --skill stop-slop-madyankin

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? AI-generated and AI-influenced prose is full of recognizable tells: throat-clearing openers, binary contrasts, dramatic fragmentation, business jargon, and formulaic rhythm. This Skill identifies and removes those patterns so text reads as natural human writing without flattening the author's distinctive voice. ## Core Features & Use Cases - Pattern Detection and Removal: Cuts filler phrases, emphasis crutches, business jargon, and formulaic structures using curated reference lists of known AI tells. - Structured Scoring: Rates prose on five dimensions (directness, rhythm, trust, authenticity, density) with a 35/50 threshold and a bounded two-pass revision loop. - Voice Preservation: Distinguishes deliberate stylistic choices from slop, flags ambiguous cases instead of silently rewriting them, and skips technical content like code, commands, and UI labels. - Use Case: A user pastes a blog draft that opens with "Here's the thing:" and leans on "Not X. But Y." constructions. The Skill rewrites it into direct statements, varies sentence rhythm, and lists any patterns intentionally left in place. ## Quick Start Ask the AI to clean up this draft so it stops sounding like AI wrote it, and paste your text.

Frequently Asked Questions about stop-slop

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

FAQPage Schema
How do I make AI-generated text sound more human?▼

Remove AI writing patterns like throat-clearing openers, binary contrasts, dramatic fragmentation, and business jargon, then vary sentence rhythm. This Skill applies curated phrase and structure lists plus a scoring rubric to rewrite prose while keeping the author's voice.

What are common AI writing tells to avoid?▼

Common tells include openers like "Here's the thing", emphasis crutches like "Let that sink in", binary contrasts like "Not X. But Y.", three-item lists, and jargon like "navigate" or "lean into". The references directory catalogs these phrases and structures.

Can it review text for AI patterns without rewriting it?▼

Yes. Requesting a review, check, or audit triggers audit-only mode: the Skill lists flagged patterns with location references, scores the text, and stops without rewriting. Ambiguous requests prompt a clarification before proceeding.

Does editing AI slop change technical content or code?▼

No. The Skill explicitly skips code blocks, filenames, CLI commands, UI labels, and technical terms embedded in prose. Instructions like "Navigate to the Settings page" are treated as UI guidance, not business jargon.

When should I not use AI slop removal on text?▼

Skip it for grammar fixes, spell-checking, general tone rewrites, SEO optimization, translation, or code review. It also does not apply when the goal is making text more formal without any concern about AI patterns.