Anti-AI-Slop Writing skill

A writing rules skill for agents: it strips model boilerplate, varies sentence rhythm and bans invented numbers and quotes

  • 577 ★ on the repository
  • Last commit
Skill

Minimal access

Why this level

  • The skill is plain text rules that run nothing and make no network calls
All reasons and checks

jalaalrd/anti-ai-slop-writing

Install

In your terminal, with SkillFoxx CLI

npx skillfoxx add skills/anti-ai-slop-writing

Detects the agents on your machine, shows the access level and pins the version.

Other ways to install

Access and checks

How we review

Minimal accessThe tool runs nothing and makes no network calls.

  • The skill is plain text rules that run nothing and make no network calls

This is third-party code. Review the repository files before installing.

What it does

Anti-AI-Slop Writing switches on for any writing task, from a tweet or an email to a report or a README description. Before working, the agent loads the ban list from references/banned-words.md and steers around it, swapping boilerplate for specifics. The list holds more than 50 words like delve, tapestry and vibrant, over 35 phrases and 16 openers such as "Certainly," and "Moreover,". Structure rules come separately. No grouping in threes and no three consecutive sentences of the same length. Paragraphs do not follow the thesis, explanation, example, transition pattern, and passive voice and swinging between "on one hand" and "on the other" are banned too. Dashes, exclamation marks and ellipses are capped per length of text, and semicolons are fine. Social posts, emails and private messages get no markdown headers, no emoji bullets and no hashtag stacks. Accuracy rules forbid inventing statistics, quotes and anecdotes, and hypotheticals are introduced with words like "imagine". When writing for a specific person, the agent matches that person's voice, and with no brief it writes directly and slightly informally, with contractions. An 11-point self-check runs before output, and the agent never mentions the rules in the text.

Who it is for. For writers, startup founders and marketers who ask Claude Code, Codex or Cursor for an English post, email or article.

Good fit when

  • You need a tweet or LinkedIn post without a hashtag stack, emoji bullets or openers like "In today's competitive landscape"
  • The agent writes a cover letter or newsletter, and you want delve and tapestry gone, uniform sentence length gone and a dash on every other line gone
  • You ask for a report or article and want the agent not to invent numbers, quotes and anecdotes
  • You want text in your voice: the agent asks whether you swear, write long or short and what you would never say
  • You prepare a README or a project description and want lively language without corporate pep

Not a fit when

  • You need text in Russian or another language: the ban lists, openers and contractions are set for English
  • You need a strict academic or legal register: the rules forbid passive voice and ask for direct speech with contractions
  • You want a scored review of finished text: the skill sets writing rules and has no separate scorer

Example request

Write a LinkedIn post about launching our app, use the anti-ai-slop-writing skill and do not invent any numbers

Limitations

  • Word lists, openers, examples and contractions are built for English text.
  • The skill limits machine-writing tells but does not test text against detectors and guarantees nothing.
  • The rules are strict and apply to any text: the ban on groups of three and on passive voice applies even where they fit.
  • Installing through the Claude Code marketplace points at another repository by the author, founder-toolkit. The files of this repository are installed by hand.

How to disable. Delete the `~/.claude/skills/anti-ai-slop-writing` folder if you installed by hand, or remove the plugin through `/plugin` if you used the marketplace. In other tools remove SKILL.md from the skills directory or from the start of the conversation.

FAQ

How do I install Anti-AI-Slop Writing in Claude Code?

With `/plugin marketplace add jalaalrd/founder-toolkit`, as the author suggests, or by hand: clone the repository and run `cp -r skills/anti-ai-slop-writing ~/.claude/skills/`. You can invoke the skill with `/anti-ai-slop-writing`, but it also fires on its own when you ask for something to be written.

Which words and phrases does the skill ban?

More than 50 words including delve, tapestry, landscape, testament, vibrant and pivotal, over 35 phrases like "In today's competitive..." and "It's worth noting...", plus 16 openers: "Certainly,", "Moreover,", "Additionally,". The agent loads the full list on demand.

How many dashes and exclamation marks does the skill allow?

One em dash per 500 words, one exclamation mark per 1,000 words and one ellipsis per piece. The author encourages semicolons.

How does the skill treat numbers and quotes?

Inventing statistics, studies and quotes is not allowed. Without a real number the agent says "roughly" or admits uncertainty, paraphrases quotes with attribution and introduces hypotheticals with words like "imagine". Real names, companies and dates are welcome.

How do I use the skill in ChatGPT, Gemini or another chat?

Copy the SKILL.md contents and paste them at the start of a conversation, where they act as a system-level writing constraint. For Cursor, Codex, Gemini CLI and other tools put the SKILL.md file in their skills directory.

Discussion

No questions or reviews yet.

Skills

A skill that edits English drafts: strips AI boilerplate, keeps the author's voice, and can also just list what it found without rewriting

SkillMinimal access13KRepository stars

A short skill with strict rules that removes stock AI phrases and structures from prose

SkillMinimal access18.5KRepository stars

Avoid AI Writing

avoid-ai-writing

A skill that finds typical machine-writing patterns in text and rewrites them, with detect-only and edit-in-place modes

SkillMinimal access5.3KRepository stars

Two skills for English prose: humanize rewrites a model draft in a natural voice, and ai-check flags machine-writing tells

SkillMinimal access587Repository stars

A skill that finds AI tells in English prose and edits them surgically; three Python scanners run on their own and the teach command learns your voice

SkillMinimal access577Repository stars

Addy Osmani skill for clearer writing: an interview and draft from your own words, a rewrite of existing text, a review that leaves the file alone

SkillMinimal access335Repository stars

Comparison with similar tools

Entries that share the main search query, and how they differ.

EntryOfficialRepository starsMain difference
Anti-AI-Slop Writing (this entry)no577
Stop Slopno18,452a short skill with strict rules against stock AI phrases in prose
sepiano3,556cleans finished text of AI tells and repairs the structure of fiction

README in short

The README states the skill's goal briefly and lists what it catches, from single words to invented facts. Then come install routes for Claude Code, other agents and chats, a two-file layout and the author's credit. The rules lean on a 2025 Carnegie Mellon study, Wikipedia's Signs of AI Writing page and Buffer's analysis of 52 million posts.

README badge

Original SKILL.md

---
name: anti-ai-slop-writing
description: Produces human-sounding text that avoids detectable AI writing patterns. Activates on any writing task — tweets, emails, articles, bios, captions, reports, copy, messages, LinkedIn posts, cover letters, README files, or any content where the output must not read as AI-generated. Enforces banned vocabulary, structural variety, punctuation discipline, accuracy rules, and voice calibration. Use when the user says "write," "draft," "rewrite," "make this sound human," "anti-slop," "not AI," or any variation of wanting authentic-sounding output.
---

# Anti-AI-Slop Writing Directive v2

Produces text that avoids statistically detectable AI writing patterns. Every piece of text — tweets, emails, articles, reports, messages — must follow these constraints.

## Before Writing Anything

Load the banned words and phrases list from [references/banned-words.md](references/banned-words.md). Never use any word or phrase on that list. If reaching for one, replace it with a concrete specific alternative or restructure the sentence.

## Structural Rules

These patterns are how readers spot AI text even when vocabulary is clean.

**No Rule of Three.** AI defaults to threes. Break it. Use two, four, one, five. Never default to three unless the content genuinely has three items.

**No uniform sentence length.** No three consecutive sentences of the same length. Ever. Mix 4-word sentences with 30-word ones. This is the single most measurable AI detection signal.

**No parataxis.** Parataxis is the AI default: short sentence. Then another. Then another. It reads like a poem and immediately signals AI authorship. Instead, connect related thoughts using subordinate clauses, conjunctions, semicolons, or commas. "Short sentence. Then another. Then another." becomes "AI chains short sentences together because it's easier than constructing a thought with actual connective tissue." Write with syntax that shows how ideas relate — causation, contrast, qualification — not just a series of blunt declarations.

**No hedging seesaw.** Pick a side. State it plainly. Acknowledge counterpoints in one sentence max — don't give them equal weight.

**No corporate pep talk tone.** Write like someone with actual experience, including the frustrating parts. No cheerleading.

**No identical paragraph structure.** AI follows: topic sentence → explanation → example → transition. Break it. Start some with questions, some with blunt statements. Let some be one sentence. Let some end without a transition.

**No excessive bullet points.** Use sparingly. Make them uneven when used — some long, some short. Never more than 5-7 in a row. If it fits in a sentence, use a sentence.

**No "As [role], I..." openers.** Real people just say the thing without announcing credentials.

**No parallel structure across sections.** Different points need different treatment. Vary section lengths.

**No passive construction.** Avoid "is being done," "was found to be," "are considered to be." Write active and direct. AI defaults to passive to sound measured; it sounds dead instead.

**Let paragraphs end abruptly.** Not every paragraph needs a summary or transition. Sometimes just stop.

## Punctuation Rules

**Em dashes:** Maximum ONE per 500 words. The single most cited AI tell in existence. Use commas, semicolons, colons, parentheses, or new sentences instead.

**Exclamation marks:** Maximum one per 1,000 words. Enthusiasm comes from word choice.

**Ellipses:** Only when genuinely trailing off. Never as transition. Max one per piece.

**Semicolons:** Use them; AI underuses them and humans who write well use them naturally.

**Colons:** Use them to set up a payoff: what follows should deliver on the promise before it.

Report a problem

Foxx AIAnti-AI-Slop Writing
Foxx AI

I am Foxx AI and I have already vetted this tool. Ask about install, setup or anything else, and I will keep it simple.