stop-slop

Detect and remove AI writing patterns from prose during drafting and editing.

18|8|Updated Jun 17, 2026
One-click install
npx skills add https://github.com/PyModel/pythinker-code --skill stop-slop-pymodel
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: stop-slop
Source: https://github.com/PyModel/pythinker-code/tree/main/.agents/skills/stop-slop
Command: npx skills add https://github.com/PyModel/pythinker-code --skill stop-slop-pymodel

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? AI-generated prose contains predictable patterns—throat-clearing openers, binary contrasts, adverb crutches, and formulaic rhythms—that make writing sound robotic. This Skill gives an LLM a concrete rulebook to catch and eliminate those tells when drafting, editing, or reviewing text. ## Core Features & Use Cases - Banned Phrase Detection: Removes throat-clearing openers, emphasis crutches, business jargon, adverbs, meta-commentary, and vague declaratives using the catalog in references/phrases.md. - Structural Pattern Rewriting: Fixes binary contrasts, negative listings, dramatic fragmentation, rhetorical setups, false agency, passive voice, and narrator-from-a-distance voice per references/structures.md. - Quality Scoring Rubric: Rates prose 1-10 across Directness, Rhythm, Trust, Authenticity, and Density, flagging anything below 35/50 for revision. - Use Case: Paste an AI-drafted blog post and ask for a slop check; the Skill flags phrases like "Here's the thing" and "Let that sink in," rewrites passive constructions, and returns cleaner prose with before/after examples from references/examples.md. ## Quick Start Ask the AI to review your draft with the stop-slop rules and rewrite any sentences that contain AI writing patterns.

Frequently Asked Questions about stop-slop

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

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

Load the SKILL.md instructions into your AI assistant and paste the draft. The Skill flags banned phrases, formulaic structures, and passive voice, then rewrites each flagged sentence directly with a named actor and specific claims.

What phrases does this AI writing checker flag?

It flags throat-clearing openers like "Here's the thing", emphasis crutches like "Let that sink in", business jargon like "lean into", all adverbs, vague declaratives like "The implications are significant", and meta-commentary announcing the essay's own structure.

How does the prose scoring rubric work?

The Skill rates text 1-10 on five dimensions: Directness, Rhythm, Trust, Authenticity, and Density. Any total below 35 out of 50 triggers a revision pass before the text is considered finished.

Can I use this skill outside Claude Code?

Yes. The README describes four setups: add the folder as a Claude Code skill, upload SKILL.md and references to a Claude Project, copy the core rules into custom instructions, or include SKILL.md in an API system prompt.

What are the limitations of rule-based AI writing detection?

The Skill targets a fixed catalog of known phrases and structures, so novel AI tics outside references/phrases.md and references/structures.md may pass through. It also enforces stylistic rules like banning all adverbs and em dashes, which may not fit every writing context.