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 agentmods add skills/darcos-loft/fluid/design-systemnpx skills add Darcos-Loft/fluid --skill design-systemgit clone --depth 1 https://github.com/Darcos-Loft/fluidWrote 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/darcos-loft/fluid/design-system)<a href="https://agentmods.dev/skills/darcos-loft/fluid/design-system"><img src="https://agentmods.dev/badge/skills/darcos-loft/fluid/design-system.svg" alt="Measured on agentmods" 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 | $0.00089 | $0.01958 |
| Opus 5 | $0.00044 | $0.00979 |
| Sonnet 5 | $0.00018 | $0.00392 |
| Haiku 4.5 | $0.00009 | $0.00196 |
Grade A, and why
design-system 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 5d 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 — 106 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Design System
A factory for brand identities. Given a short brief, it produces a complete, unique design system in the house DNA: a token file, a Tailwind config, a brand rules doc the agent obeys, a storybook, and a human spec. The opposite of a template: the DNA (method and quality bar) stays constant, the output (palette, type, shape, mood) is different every time.
It is the piece that kills the generic look at the source. The fluid skill handles motion, the detector catches slop after the fact, this gives each project an identity from line one.
Built from four hand-made systems: MVP "Liquid Glass" (the structural template), LeCoffre (the delivery format), PLR @plr/ui, and Vigie.
When to use
- Starting a new client site or product and you need its visual identity defined.
- Producing or extending design tokens (color, type, spacing, radii, shadows, motion).
- You want a
tokens.css+ Tailwind config + brandCLAUDE.md+ storybook for a project. - An existing UI feels generic and needs a committed identity.
The two halves
A system here is DNA (constant) plus brand (variable). Get both right.
- The DNA is the house method and quality bar. It never changes. See
reference/dna.md. - The brand is what makes this system unlike any other: palette, type, shape, surface, mood. Derived per project from the brief. This is where you must NOT converge.
init: the workflow
-
Gather the brief (ask only what is missing):
- Brand name and sector/positioning.
- Personality, pick an extreme: institutional/refined, editorial, luxury, playful/energetic, techy/precise, brutalist/raw, organic/warm. Half-measures read as generic.
- Light, dark, or both (auto).
- Surface treatment: flat, soft (subtle shadow/relief), or glass (liquid, translucent).
- One anchor or constraint: an existing logo color, a reference you admire, a competitor you must NOT resemble.
-
Derive the brand variables (commit, do not hedge):
- Start from the database: pick a palette from
data/palettes.md(100+ committed palettes), a type pairing fromdata/type-pairings.md(50+ distinctive pairings), a style direction fromdata/styles.md, and a layout archetype fromdata/layouts.mdthat fit the brief. Adapt the exact values; never improvise a generic default, and never default to the split-hero-plus-three-cards skeleton. Two clients in the same sector get different picks, including a different page shape. For motion reach intodata/motion-presets.md, for any data display intodata/chart-types.md, and for the build pick a stack fromdata/tech-stacks.md. - Color: one dominant brand hue + one sharp accent + semantic (success/warn/danger) + tuned neutrals. Never pure black. See
reference/color.md. - Type: a distinctive display font paired with a refined body font, self-hosted. Never Inter/Roboto/Arial as the primary. See
reference/typography.md. - Shape: a radii scale (sharp 4 to 8, rounded 12 to 22, or pill) chosen to match the personality.
- Surface/relief: flat, soft-shadow, or glass tokens.
- Motion: the house easings, already in the template (
--ease-out,--spring,--ease-in-out), shared withfluid.
- Start from the database: pick a palette from
What ships with it
17 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.
- data/chart-types.md 11 KB
- data/layouts.md 3.7 KB
- data/motion-presets.md 4.2 KB
- data/palettes.md 21 KB
- data/styles.md 19 KB
- data/tech-stacks.md 14 KB
- data/type-pairings.md 8.7 KB
- reference/color.md 1.8 KB
- reference/dna.md 2.9 KB
- reference/glass.md 3.7 KB
- reference/icons.md 3.0 KB
- reference/rules.md 2.8 KB
- reference/typography.md 1.9 KB
- templates/CLAUDE.md 2.5 KB
- templates/storybook.html 3.1 KB
- templates/tailwind.config.js 1.5 KB runs code
- templates/tokens.css 7.1 KB
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.
- 5d ago First seen · 106 lines · 89 tokens per session scan A 0ffbbebcf325
design-system is a skill published in the GitHub repository Darcos-Loft/fluid (3 stars, last pushed 1mo ago), licensed MIT. It adds 89 tokens to every session and 1,958 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
tailwind-design-system
Build scalable design systems with Tailwind CSS v4, design tokens, component libraries, and responsive patterns. Use when creating component libraries, implementing design systems, or standardizing UI patterns.
designlang-tokens
Use when styling UI for cal.com — references the extracted design system tokens instead of inventing colors, spacing, or typography.
paint
Paint a complete visual universe with genjutsu - art direction brainstorm, design system, implementation, audit. Anti-AI-slop design pipeline. Adapts to Web, Android (Compose), Apple (SwiftUI).
component-family-consistency
Buttons, inputs, pills, badges, calendars, and other interactive components form a visual family — they share the same border-radius, colour logic, shadow scale, border style, and spacing rhythm. Inconsistency between them breaks the sense of a coherent product. Use when building or reviewing a component library…
extract-design
Extract a complete design system — colors, typography, spacing, components, shadows, and W3C design tokens — from any live website using Dembrandt. Runs a headless browser against the URL and returns real computed values from the DOM. Use when you need a site's actual design tokens, want to reverse-engineer a visual…
data-display-and-selection
Complex data deserves multiple view modes — grid, list, table — chosen by the user based on their task. Row and item selection should use large hit areas (the whole row or card, not just a checkbox). Selected state is communicated through a subtle background colour shift. Mass actions appear when items are selected.…