stop-slop

what it does

Remove AI writing patterns from prose. Use when drafting, editing, or reviewing text to eliminate predictable AI tells.

by Hardik Pandya · Hardik Pandya, MIT · version 1 · license MIT

Verified Before and after: 2 models 1 try per case View source Download .zip

Compare the answers

Answers from Claude Sonnet 5.5 and GPT-6.1 Sol · try 1 of 1

Same request, same model, once with the skill and once without. Read both and judge for yourself.

“Tighten this paragraph for our company blog.”

Source material given with the request

In today's fast-moving landscape, it's worth noting that onboarding isn't just a checklist — it's the foundation of everything. Here's the thing: new hires who get a clear first week are, quite frankly, significantly more likely to stay. At Brightloop, we navigate this by pairing every new hire with a buddy for their first 30 days, and honestly, the results speak for themselves: 91% of 2025 hires were still with us after a year.

with the skill

New hires who get a clear first week stay longer. At Brightloop, we pair each new hire with a buddy for their first 30 days, and 91% of our 2025 hires stayed past their first year. I cut "significantly more likely" because the paragraph gives no figure to support it. If you have a comparison, such as retention before the buddy program or the industry average, add it next to the 91%. It will persuade more than any adjective.

without

**Tightened version:** Onboarding isn't a checklist. It's the foundation for everything that follows. New hires who get a clear first week are significantly more likely to stay. At Brightloop, we pair every new hire with a buddy for their first 30 days, and 91% of our 2025 hires were still with us a year later. **What changed:** - Cut the filler openers and intensifiers ("In today's fast-moving landscape," "it's worth noting," "Here's the thing," "quite frankly," "honestly"). - Replaced "navigate this by" with a direct statement of what you do. - Dropped "the results speak for themselves" and let the 91% stat do that work. - Swapped the em dash for a period to give the opening more punch. If you want it even shorter, you could trim the second sentence: *"Onboarding isn't a checklist; it's the foundation. New hires with a clear first week are far more likely to stay."*

Answers are shown exactly as the model wrote them, from the first try, and nothing is picked by hand.

The models

2 models

Claude Sonnet 5.5

made by Anthropic

claude-sonnet-5-5 · revision not reported by the provider

1 try per case · tested 1 Nov 2026

GPT-6.1 Sol

made by OpenAI

gpt-6.1-sol · revision not reported by the provider

1 try per case · tested 1 Nov 2026

The tasks

written by Carte Blanche

essay-opening

“Write the first two paragraphs of a blog post arguing that teams under ten people should stop holding daily standups.”

explain-retries

“Explain to a new engineer why our service retries failed webhook deliveries with exponential backoff. Two short paragraphs.”

launch-post

“Write a LinkedIn post announcing that we shipped offline mode for our note-taking app. About 150 words.”

price-change-email

“Write a short email telling our customers we're raising the Pro plan price, and why.”

tighten-blog-paragraph

“Tighten this paragraph for our company blog.”

Install

Download the .zip and unzip it. Put its files, in a folder named stop-slop, here:

Claude Code~/.claude/skills/Every project. For one project only, use .claude/skills/ inside it.

Codex~/.agents/skills/Every project. For one repository only, use .agents/skills/ inside it.

Cursor~/.cursor/skills/Cursor also reads the Claude Code and Codex folders above.

Then start a new session and ask for what the skill does. Other agents that read SKILL.md have their own skills folder; check their docs.

Coming after launch: get skills from inside your agent, no download needed.