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/mhylle/claude-skills-collectionnpx agentmods add skills/mhylle/claude-skills-collection/design-languageWrote 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/mhylle/claude-skills-collection/design-language)<a href="https://agentmods.dev/skills/mhylle/claude-skills-collection/design-language"><img src="https://agentmods.dev/badge/skills/mhylle/claude-skills-collection/design-language/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/mhylle/claude-skills-collection/design-language"><img src="https://agentmods.dev/badge/skills/mhylle/claude-skills-collection/design-language.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 | $0.00260 | $0.03685 |
| Opus 5 | $0.00130 | $0.01843 |
| Sonnet 5 | $0.00052 | $0.00737 |
| Haiku 4.5 | $0.00026 | $0.00368 |
Grade A, and why
design-language scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 263 lines — stays where its author put it; the contents beside it link to each section on GitHub.
design-language — a design-brief → living-styleguide compiler
You help the user design and manufacture their own reusable visual design language and
emit it as a self-contained ./design-system/ folder that future Claude sessions read and
build from. You are not styling one screen — you are compiling a durable language.
The whole point is to land somewhere that is genuinely the user's own, deliberately away from the default Anthropic/Claude aesthetic. Read references/anthropic-anti-reference.md so you know exactly what you are steering away from and can detect regression toward it.
SKILL_DIRbelow means the directory containing this SKILL.md. The scripts import a sharedtokens_libsibling, so always invoke them by their real path (python3 "$SKILL_DIR/scripts/<name>.py" …) — Python puts the script's own dir on the path, so the import resolves no matter what the user's working directory is. The./design-system/output is created in the user's current working directory.
The non-negotiables (read once, hold throughout)
./design-system/css/tokens.cssis the single source of truth. Every dashboard page and the W3C JSON are regenerated from it. Never hand-author a page; change tokens, then regenerate. (Invariant from the spec — it is why the visual styleguide can never drift from the machine-readable tokens.)- Reaction-first. Don't interrogate the user with abstract taste questions up front. Show them things — researched real references, and live archetypes rendered in their own emerging tokens — let them react, then refine with targeted questions. People know what they like when they see it far better than they can specify it cold.
- Real divergent thinking, every session. The experimental end of each dimension's spectrum must be reasoned freshly from this user's north star. The spinning 3D image cloud in the imagery seed is one illustration of thinking outside the box — it must NOT become the answer every time. If every language you help build ends with the same experimental gimmick, you have failed the core promise. See "Authoring the spectrum".
- Phasing is for resumability, never scope-cutting. The v1 scope is the entire system. Never quietly drop a dimension to finish faster.
- Portable output. Everything under
./design-system/uses relative paths only, no CDNs, works offline — so the folder can be moved, zipped, and committed to any repo.
What ships with it
24 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- agents/design-researcher.md 8.4 KB
- assets/dashboard/base.css 4.1 KB
- assets/dashboard/components.css 7.3 KB
- assets/dashboard/demos.js 4.2 KB runs code
- assets/questionnaire.html.tmpl 32 KB
- evals/evals.json 6.2 KB
- references/anthropic-anti-reference.md 7.9 KB
- references/contract-format.md 7.9 KB
- references/question-types.md 12 KB
- references/token-schema.md 8.8 KB
- scripts/assemble_dashboard.py 4.2 KB runs code
- scripts/dashboard/__init__.py 442 B runs code
- scripts/dashboard/assets.py 1.3 KB runs code
- scripts/dashboard/brief.py 841 B runs code
- scripts/dashboard/contract.py 3.5 KB runs code
- scripts/dashboard/pages_compare.py 3.3 KB runs code
- scripts/dashboard/pages_overview.py 6.4 KB runs code
- scripts/dashboard/pages_showcase.py 13 KB runs code
- scripts/dashboard/pages_tokens.py 4.9 KB runs code
- scripts/dashboard/render.py 3.3 KB runs code
- scripts/init_state.py 11 KB runs code
- scripts/render_spectrum.py 37 KB runs code
- scripts/serve_questionnaire.py 22 KB runs code
- scripts/tokens_lib.py 15 KB runs code
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 · 263 lines · 260 tokens per session scan A f6bedef5e79f
design-language is a skill published in the GitHub repository mhylle/claude-skills-collection (18 stars, last pushed 8d ago), licensed MIT. It adds 260 tokens to every session and 3,685 once invoked, about $0.0013 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
chakra-ui-builder
Build responsive, accessible UI components and layouts using Chakra UI v3, install or configure Chakra UI in new and existing projects, and design scalable themes using tokens, semantic tokens, recipes, and slot recipes. Use this skill whenever a user asks to build, create, or generate any UI component, page, form…
visual-ralph
Visual Ralph orchestration for frontend UI from generated references, static references, or live URL targets, using $ralph with built-in visual verdict and pixel-diff evidence until the implementation matches and leaves a reproducible design system.
frontend-ui-dark-ts
Build dark-themed React applications using Tailwind CSS with custom theming, glassmorphism effects, and Framer Motion animations. Use when creating dashboards, admin panels, or data-rich interfaces with a refined dark aesthetic.
reveal-hover-effect
Build cursor-following spotlight reveals that expose a second aligned image through a soft radial mask. Use for hover-to-color, before-and-after, x-ray, material, texture, product-detail, and illustrated hero effects where a desaturated or embossed base image should remain visible while another treatment follows an…
frontend-visual-qa
Audits already-rendered web, landing-page, HTML deck/slide, browser tool/game, dashboard/admin, design-system, and desktop UIs using real-browser or native-app journeys, inspected screenshots, DOM geometry, responsive or projection viewports, and a bundled Playwright sweep. Use after UI implementation to find…
prototype-web
A clickable, high-fidelity web product prototype with navigation, a hero section, feature cards, steps, social proof, and optional pricing. It is designed to resemble a finished landing page while remaining a prototype.