xiry.

AI · GitHub repo

stop-slop

A skill file and prompt system for removing AI writing patterns

From the Xiry library · Updated 10 October 2026

Visit github.com ↗

stop-slop is an open-source Claude skill and prompt reference that detects and removes predictable AI writing habits from prose. It provides clear rules against throat-clearing intros, cliché contrast formulas, false agency, and robotic rhythms to make generated text sound natural. It can be installed as a Claude Code skill, added to Claude Project knowledge, or used directly in system prompts.

What it's for

  • Clean up AI-generated blog posts and essays to remove repetitive phrasing
  • Integrate into Claude Projects as persistent knowledge for natural-sounding copy
  • Use as system instructions in LLM API workflows to enforce active voice and concision
  • Score writing quality across dimensions like directness, rhythm, and density

Features

  • Core instruction file (SKILL.md) compatible with Claude Code and LLM system prompts
  • Banned phrases reference cataloging throat-clearing, filler adverbs, and business jargon
  • Structural guide identifying binary contrasts, staccato fragmentation, and false agency
  • Before-and-after transformation examples illustrating clean edits
  • 5-dimension scoring rubric (Directness, Rhythm, Trust, Authenticity, Density) to evaluate text

Pricing

Open source. Free under the MIT license

Details

Type
GitHub repo
Category
AI
Topic
Prompt Engineering
Open source
Yes
Platforms
Command line, API, Web, other
For
Writers, developers, and marketers using Claude or other LLMs to write copy and prose.
Tags
claude skill, writing, copywriting, llm prompts, editing, claude

Why it was saved

  • @tom_doerr, 24 May 2026: Teaches Claude (or any LLM) to catch and remove predictable AI phrases, structures and rhythms from writing.

Related

All 18 in Prompt Engineering →

Updated