All directories
SkillMarketingDesign

Anti-Slop

Rand’s comprehensive anti-slop toolkit for text, code, and design — detect_slop / clean_slop scripts plus pattern libraries across natural language, programming, and UI.

Open on GitHubBrowse more skills
01

Overview

Anti-Slop (rand/cc-polymath, also published on Smithery, LobeHub, AIMCP, and MCP Market) is a multi-domain quality skill. It detects generic AI patterns in natural language, code, and design, scores text slop 0–100, and offers automated cleanup for high-risk phrases with preview/backup safety. Reference packs cover text-patterns, code-patterns, and design-patterns. Use it when you need one skill that can review a blog post, a PR full of `data`/`result` variables, and a purple-gradient landing page.

SaaSAgencyOther

Work with LimeDock

Using this skill? LimeDock can wire it into a durable automation you own.

Skills show what's possible. LimeDock builds and runs the owned marketing, sales, and ops automations around them.

We sell owned automations for SaaS teams — live workflows that plug into Slack, CRM, and your internal platform — not just a skill list.

Book a workflow call
02

Link

https://github.com/rand/cc-polymath
03

Installation guide

npx skills add rand/cc-polymath -a claude-code
# or
npx -y @lobehub/market-cli skills install rand-cc-polymath-anti-slop --agent claude-code

Skill path in repo: `skills/anti-slop`. Scripts: `python scripts/detect_slop.py <file>` and `python scripts/clean_slop.py <file> [--save]`.

04

How to use it

1. Detect: run detect_slop.py or ask the skill to score a file. 2. Clean text with clean_slop.py in preview mode first. 3. For code, follow code-patterns.md (rename generics, delete obvious comments). 4. For design, audit against design-patterns.md (gradients, card spam, generic CTAs). 5. Always manually review aggressive cleanups.

05

Example prompts

  • “Use Anti-Slop to detect_slop on docs/launch.md --verbose and summarize findings.”
  • “Clean this blog draft with anti-slop, then show a diff of removed phrases.”
  • “Review this React page for design slop patterns and propose concrete alternatives.”
  • “Help me set a team policy: reject docs with slop score > 40.”
06

Use cases and examples

  • Score a documentation file for text slop before release
  • Clean buzzwords and meta-commentary from AI-drafted docs
  • Refactor generic AI code names and needless wrappers
  • Design review for template-like SaaS UI patterns
  • Establish team max slop-score thresholds in CI
07

Prerequisites

  • Python available if you want the detect/clean scripts
  • Claude Code / Cursor / Codex with skills support
  • Judgment — not every hedge or pattern is slop in legal/academic contexts
08

Tips

  • Scripts handle text; code and design cleanup stay mostly manual with the reference guides.
  • Preview before --save; backups are created when overwriting.
  • Same skill appears under several marketplace IDs — install once from rand/cc-polymath.
09

Sources