Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/Abdulkader-Safi/AI-Writing-Rulesnpx agentmods add skills/abdulkader-safi/ai-writing-rules/anti-ai-writingWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/abdulkader-safi/ai-writing-rules/anti-ai-writing)<a href="https://agentmods.dev/skills/abdulkader-safi/ai-writing-rules/anti-ai-writing"><img src="https://agentmods.dev/badge/skills/abdulkader-safi/ai-writing-rules/anti-ai-writing/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/abdulkader-safi/ai-writing-rules/anti-ai-writing"><img src="https://agentmods.dev/badge/skills/abdulkader-safi/ai-writing-rules/anti-ai-writing.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00078 | $0.01823 |
| Opus 5 | $0.00039 | $0.00911 |
| Sonnet 5 | $0.00016 | $0.00365 |
| Haiku 4.5 | $0.00008 | $0.00182 |
Grade A, and why
anti-ai-writing scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 144 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Anti-AI writing
Writing that reads like a sharp, plain-spoken human. Never like AI.
The one-line test
If you can cut a word and keep the meaning, cut it.
How to use this
Writing something new: read the hard rules below, write, then run the checklist at the end.
Editing existing text: run /deslop <file> or work through the checklist against the draft.
Diagnosing why something reads as AI: find the pattern in the table and open its file in research/.
Hard rules
No em dashes. Not a density target, a ban. Use a full stop, a comma, a colon, or brackets. A hyphen inside a word like "plain-spoken" is fine. See research/15-em-dash-density.md.
Straight quotes only. No curly quotes, no apostrophe character, no ellipsis character. Three dots, or better, no ellipsis.
Sentence case headings. Capital on the first word only. Never Title Case. Never skip a heading level. Never open a document with a heading that repeats its own title.
No negative parallelism. Kill "not just X, it's Y", "it's not X, it's Y", "not only X but also Y". Delete the negation and keep the positive half. If the positive half cannot stand alone, the sentence had no content.
Break the rule of three. Count the items in every list. If they are all three, use one, two, or four instead. This applies to bullets, features, steps, and adjective strings.
Bold sparingly. At most one thing per screen. No stacked **Label:** text bullets.
No wrap-up ending. No "in conclusion", no "challenges and future prospects", no "as we look ahead", no closing paragraph that restates the opening. End at the last real sentence.
Name the source or drop the claim. No "experts say", no "studies have shown", no "research suggests".
No assistant scaffolding. No "Certainly!", no "I hope this helps", no "As an AI", no placeholder brackets, no knowledge-cutoff notes.
Vocabulary
Avoid: delve, tapestry, vibrant, multifaceted, nuanced, intricate, realm, landscape (figurative), testament, navigate (figurative), foster, leverage, robust, seamless, underscore, pivotal, crucial, comprehensive, holistic, transformative, elevate, embark, unlock, harness, showcase, resonate, garner, boasts, meticulous, bolster, streamline, empower, myriad, plethora, paradigm, interplay, groundbreaking, cutting-edge, world-class, enterprise-grade.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 144 lines · 78 tokens per session scan A 3a5476a42c0f
anti-ai-writing is a skill published in the GitHub repository Abdulkader-Safi/AI-Writing-Rules (30 stars, last pushed 23d ago), licensed MIT. It adds 78 tokens to every session and 1,823 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
humanize-pro
Use when the user asks to humanize, unslop, de-slop, de-AI, clean up, tighten, or make text sound human or natural; when they say something reads like AI or like ChatGPT; when they ask whether something matches their voice or style guide; or when drafting or rewriting outbound human communication: LinkedIn posts and…
anti-slop
Remove AI slop from any voice-bearing prose — original posts, threads, articles, long-form, emails, docs, READMEs, marketing copy, bios, scripts. Use when drafting text meant to sound like a specific person or brand, when rewriting text that reads generic, corporate, or AI-generated, or when asked to humanize…
kill-slop
Audit a file or draft for AI slop and off-voice lines against the author's bound voice, report findings with in-voice swaps, then apply approved fixes. Use when the user says "kill slop", "/kill-slop [file]", "find the slop in this", "audit this for slop", or wants a deck, article, page, or draft cleaned to sound like…
cco-tools
Show what tools actually cost in tokens — learned per-tool averages from observed results, replacing the hardcoded MCP/Agent guesses.
test-run
Run plugin test suites in this monorepo and report a concise pass/fail summary. Optional plugin slug arg; without arg, runs all plugins under plugins/.
github-pr-creation
Creates GitHub Pull Requests with automated validation and task tracking. Use when user wants to create PR, open pull request, submit for review, or check if ready for PR. Analyzes commits, validates task completion, generates Conventional Commits title and description, suggests labels. NOTE - for merging existing…