跳至正文
OpenAgentSkill
浏览已开放 · 目录记录日期 2026-11-21 · 自动更新和账户服务暂时关闭。
开发与测试

The Antislop

A Claude Code skill that detects and fixes AI-generated writing patterns with tiered scoring and an editor mode.

aplaceforallmystuff · 已有 AI 审核记录

安装与使用

运行命令前先阅读来源与依赖说明。复制命令不会自动安装或执行。

npx skills add aplaceforallmystuff/the-antislop

技能说明

The Antislop

Detect and fix AI-generated writing patterns (slop).

A Claude Code skill that detects and fixes AI-generated writing patterns. More comprehensive than detection alone — it actually rewrites the problems.

Why

AI-generated text has tells. Wikipedia's Signs of AI Writing documents 24 patterns. But detection alone isn't enough. You need a tool that:

  1. Detects comprehensively — 35+ patterns across content, language, style, and structure
  2. Scores severity — Not all patterns are equally bad (tiered system)
  3. Actually fixes problems — Editor mode rewrites, not just flags
  4. Catches structural tells — Staccato fragments, manufactured personality, uniformity

Install

# In Claude Code:
/plugin marketplace add aplaceforallmystuff/marketplace
/plugin install the-antislop@aplaceforallmystuff
mkdir -p ~/.claude/skills
git clone https://github.com/aplaceforallmystuff/the-antislop.git ~/.claude/skills/the-antislop

Use cases

  • Use it when a draft "sounds like AI" and you want the specific tells named and fixed.
  • Use it as a final pass before publishing any AI-assisted content.
  • Use it to humanize AI-generated text — cut Tier 1 phrases, break up staccato fragments, vary sentence rhythm.
  • Use it to check a draft against the Horoscope Test before it goes out under your byline.
  • Use it to audit without editing — request "audit only" when you just want the slop score.

How it works

  1. Run the Horoscope Test — Could anyone have written this for anyone?
  2. Scan for patterns — 45+ known AI tells across 6 categories
  3. Calculate slop score — Tiered s