Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add whatsuppiyush/make-my-website-non-ai-looking-claude-skill --skill make-websitegit clone --depth 1 https://github.com/whatsuppiyush/make-my-website-non-ai-looking-claude-skillWrote 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/whatsuppiyush/make-my-website-non-ai-looking-claude-skill/make-website)<a href="https://agentmods.dev/skills/whatsuppiyush/make-my-website-non-ai-looking-claude-skill/make-website"><img src="https://agentmods.dev/badge/skills/whatsuppiyush/make-my-website-non-ai-looking-claude-skill/make-website/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/whatsuppiyush/make-my-website-non-ai-looking-claude-skill/make-website"><img src="https://agentmods.dev/badge/skills/whatsuppiyush/make-my-website-non-ai-looking-claude-skill/make-website.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00074 | $0.10121 |
| Opus 5 | $0.00037 | $0.05060 |
| Sonnet 5 | $0.00015 | $0.02024 |
| Haiku 4.5 | $0.00007 | $0.01012 |
Grade A, and why
make-website 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 11d 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 — 760 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Make a website that does not look AI-made
Everything needed to take a site from a one-line brief to something shippable: how to read the brief, how to choose a direction, how to build each layer, and the thirty patterns that will give the whole thing away if you let them through.
Two halves. Taste decides what the page looks like. Tells decide what it must never look like and what it must contain. Neither works alone. Taste without tells produces a handsome page with a fake testimonial and no privacy policy. Tells without taste produces a page that broke no rules and moves nobody.
0. The loop
Follow this order. Most bad output comes from skipping straight to step 5.
- Read the brief. One line: what, for whom, what feeling. §1
- Set the dials. Variance, motion, density. §2
- Pick the foundation. Stack, and whether a real design system applies. §3
- Pick the direction. One of five aesthetic families, chosen not defaulted. §4
- Build, in this order: type → colour → layout → images → states → motion. §5 to §10
- Build the trust surface. Demo, proof, legal. §11
- Spend the tell budget deliberately. §12
- Pre-flight. §14
If the user said only "build me a website", do not ask five questions. Infer, state your read in one line, name your assumptions, and build. Ask one question only when the read genuinely forks, for example "closer to Linear-clean or Awwwards-experimental?"
If the user says "make it not look AI-generated" and nothing else, that is a brief. It means: real images, real copy, a real trust surface, one committed aesthetic, and none of §12 Class A. Build to that and say what you assumed.
1. Read the brief
Before any code, work out what is actually wanted. The most common failure is jumping to a default aesthetic instead of reading the room.
Signals to read.
- Page kind. Landing (SaaS, consumer, agency, event), portfolio (developer, designer, studio), editorial, app shell, docs.
- Vibe words they used. Minimalist, calm, Linear-style, Awwwards, brutalist, premium, Apple-y, playful, serious B2B, editorial, glassy, dark tech.
- References. URLs, screenshots, competitors, products they named. These outrank your taste.
- Audience. A procurement panel, a design-conscious consumer, a recruiter scanning for fifteen seconds. The audience picks the aesthetic.
- Existing brand. Logo, colour, type, photography. If any exists it is starting material, not optional input.
- Quiet constraints. Accessibility-critical, public sector, regulated, children, trust-first commerce. These override aesthetic preference entirely.
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.
- 11d ago First seen · 760 lines · 74 tokens per session scan A 0f070cc9d8a6
make-website is a skill published in the GitHub repository whatsuppiyush/make-my-website-non-ai-looking-claude-skill (2 stars, last pushed 26d ago), licensed MIT. It adds 74 tokens to every session and 10,121 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 2026-08-31.
Other skills, from other repositories
avoid-ai-design
Audit and rewrite frontend UI to remove generic AI design patterns ("AI slop"). Use this skill when asked to "de-slop a UI", "make a design look less AI-generated", "audit a component or page for AI design tells", or "fix Claude/Codex-generated frontend that looks generic". Covers HTML/CSS and React/Tailwind/shadcn.…
anti-slop-design
Eliminates generic AI-generated design ("AI slop") from any front-end you build or review — web or mobile. Use whenever generating, reviewing, refactoring, or critiquing a website, landing page, dashboard, web app, iOS or Android app, React Native or Flutter screen, component, design system, or micro-interaction — and…
json-render-ui
Generate guardrailed UI from natural language. Emits constrained JSON that a Preact runtime renders. Use when the request is for a dashboard with metrics, charts, or tables; an admin panel; a data visualization interface; or a form-based application.
web-design
A web design workflow for turning a brief, reference website, screenshot, or keywords into a design specification and then web code. It covers visual style, page structure, interaction, responsive layouts, and motion.
web-design-guidelines
UI/UX best practices and accessibility audit. Use when reviewing UI code, checking accessibility, running accessibility audits, auditing forms, or ensuring web interface best practices. Triggers on "audit accessibility", "check WCAG", "review UI", "check accessibility", "audit design", "review UX", or "check best…
cinematic-scroll
Design and build cinematic websites, 3D websites, interactive portfolios and product landing pages with scroll-driven storytelling, parallax, text animation and optional Three.js/WebGL scenes. Use for new experiences, motion improvements, scroll audits and storyboards in standalone HTML or an existing app. Includes…