Liquid Glass Frontend logo

Liquid Glass Frontend

Community
ngocanhnckh
liquid-glass-frontend

Use when building a beautiful, distinctive, awwwards-tier web frontend — landing pages, dashboards, web apps, marketing sites — that needs Apple-style liquid-glass surfaces, choreographed motion.dev animations, and creative editorial layouts instead of the generic SaaS look (icon-card rows, centered heroes, default fonts, uniform fade-ins). Covers creative direction and signature interactions, the frost/blur glass material, single-source color tokens with dark+light theming, scroll/stagger/parallax/scroll-scrubbed animation, type-as-hero typography, and asymmetric grid-breaking composition.

Overview

Publisherngocanhnckh
Repositoryliquid-glass-frontend-skill
Skill nameliquid-glass-frontend
Stars
10
Forks
1
Bundled files
10
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 10 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by ngocanhnckh on GitHub. Read the source before you install it.

Installation

Install the Liquid Glass Frontend AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/ngocanhnckh/liquid-glass-frontend-skill/tree/main
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/ngocanhnckh/liquid-glass-frontend-skill.git \
  .claude/skills/liquid-glass-frontend
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Liquid Glass Frontend in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Liquid Glass Frontend on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Liquid Glass Frontend is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Liquid Glass Frontend

Build frontends people remember: real Apple-style frosted glass, motion that feels alive, layouts that break the grid on purpose, and a token system you re-skin in three numbers. This skill is the distilled, production-tested playbook behind a full enterprise app + landing site.

The bar is an Awwwards Site of the Day — not a template. Editorial confidence, a signature interaction, distinctive type, and intentional asymmetric composition. Never the generic SaaS vocabulary of three icon-cards in a row, a centered hero, default fonts, and uniform fade-ins. If what you're building looks like every other landing page, it's not done. references/creative-direction.md is the taste layer — read it.

Announce at start: "I'm using the liquid-glass-frontend skill to design this interface."

The non-negotiables

These are what separate a memorable interface from generic AI output. Hold all of them:

  1. Aim for awwwards-tier, not template-tier. Decide the ONE signature moment someone will screenshot, and over-invest there. Kill the generic component vocabulary (icon-card rows, centered hero, even spacing). See references/creative-direction.md.
  2. Commit to one bold aesthetic direction. Pick an extreme (warm editorial, refined minimal, retro-futuristic, brutalist, luxury) and execute it precisely. Timid middle-ground reads as slop.
  3. Typography is the design. Big, characterful, mixed — italic accents, variable-font axes, mono kickers, oversized numerals. No default fonts (Inter/Roboto/Arial/system-ui as the character face), no purple-on-white gradient. See references/typography-color.md.
  4. Break the grid deliberately, and vary the treatment. Asymmetric bands, overlap, layered depth, dramatic scale contrast, generous negative space. Not rows of equal cards — and not the inverse trap of making every section the same uniform linear list. Sibling items vary in size/proportion/density (bento, staggered cluster); adjacent sections contrast. Never repeat one uniform treatment. See references/layout-composition.md.
  5. Choreograph the motion. A signature interaction (masked text reveal, scroll-scrubbed motion, cursor-reactive aura, color chapters) + orchestrated page-load + scroll reveals — all sharing one easing signature. Not fade-in-up on everything. See references/motion.md and references/creative-direction.md.
  6. Liquid glass means real frost, not transparency. A see-through rgba() box is NOT glass. Glass = heavy backdrop-filter: blur() + saturate() + a specular top edge + a colored aura behind it so the frost has something to refract. See references/liquid-glass.md.
  7. One source of truth for color. The entire palette derives from three RGB triplets. Re-skinning is editing three lines. See references/theming.md.

Workflow

  1. Decide the aesthetic + the signature moment — tone, audience, and the ONE thing someone will screenshot. Write both down in a sentence before any code. (references/creative-direction.md)
  2. Lay the token foundation — drop in assets/globals.css, set the three brand triplets, pick characterful fonts. This makes everything else cohesive by default.
  3. Build the atmosphere first — background gradient mesh / blobs / grain. Glass needs something colorful behind it to read. A glass card on flat #fff is invisible.
  4. Compose editorially — asymmetric bands, type-as-hero, layered depth, dramatic scale contrast. Kill the icon-card-row instinct. (references/layout-composition.md + references/creative-direction.md)
  5. Skin surfaces in glass — .glass-surface for panels, .glass-inset for tiles-on-glass, .glass-menu for popovers, .glass-btn for buttons.
  6. Choreograph motion — build the signature interaction, then page-load stagger, then scroll reveals, then micro-interactions — all on one easing curve. references/motion.md has copy-paste variants.
  7. Add the light theme — the tokens already support it; verify glass still reads (light mode needs thinner, more saturated frost — see the gotcha below).
  8. Run the honesty check — would it place on awwwards, or does it look like every SaaS page? If the latter, push type/layout/motion harder. (references/creative-direction.md)

Tech baseline

  • Next.js (App Router) + React + TypeScript — but the CSS system is framework-agnostic; the glass utilities are plain CSS.
  • Tailwind CSS v4 — @theme for tokens, @utility/plain classes for the glass material. (v3 works too; put the utilities in a @layer.)
  • motion (motion/react, formerly framer-motion) — the animation engine. import { motion, useScroll, useTransform } from "motion/react".
  • Fonts via next/font/google (or @fontsource). Pick characterful — examples in references/typography-color.md.

The five glass utilities (what to reach for)

ClassUse forHas backdrop-filter?
.glass-surfaceLarge panels: sidebars, nav bars, hero cardsYes — heavy frost
.glass-surface-softFloating elevated panels needing more depthYes — heaviest
.glass-menuDropdowns / popovers over busy contentYes, but ~94% opaque so text stays readable
.glass-insetTiles, chips, list items inside a glass panelNo — see the #1 gotcha
.glass-inset-hoverHover state for tabs/buttons/menu itemsNo — sheen lift, not a color swap
.glass-btn (+--clay/--ink)ButtonsYes

Top gotchas (learned the hard way)

  • Never nest backdrop-filter inside backdrop-filter. It breaks compositing — the inner frost renders garbage or nothing. A glass panel frosts the background once; everything inside it uses .glass-inset (a translucent tint + top sheen, no filter of its own).
  • Glass is invisible without a colored backdrop. Put a gradient mesh / blob / aura behind glass surfaces. On a flat solid color there is nothing to refract and it looks like a plain semi-transparent box. The --aura token controls this.
  • Light mode needs the opposite tuning from instinct. White panels hide the frost. In light theme, use a thinner panel mix (~30% instead of 42%) and higher saturate() (210%+) so the pastel aura reads through. The provided CSS already does this — keep the override.
  • Popovers get trapped behind content. An ancestor with backdrop-filter creates a stacking context that traps z-index. Render dropdowns/modals at the shell root with position: fixed and a high z, not nested where they're declared.
  • Tailwind v4 @theme caches aggressively. After editing tokens, restart the dev server with a clean cache (rm -rf .next) — HMR won't always pick up @theme changes.

Reference files (load when you need depth)

  • references/creative-direction.md — read first. The awwwards-tier taste layer: the signature-moment rule, the generic-vocabulary kill list, type-as-hero, signature interaction patterns, the honesty check.
  • references/liquid-glass.md — the full material system, every utility explained, layering rules, the colored-aura technique.
  • references/motion.md — page-load orchestration, scroll reveals, stagger, parallax, animated gradients, the house easing curve.
  • references/theming.md — three-triplet palette, derived tokens, dark/light surface tokens, opacity composition.
  • references/typography-color.md — font pairing that isn't slop, choosing a palette, the "avoid generic AI aesthetics" rules.
  • references/layout-composition.md — asymmetric bands, overlap, grain, geometric line-art, spacing rhythm.

Drop-in assets

  • assets/globals.css — the complete stylesheet: brand tokens, glass utilities, theming, atmosphere. Copy it, set three triplets, go.
  • assets/examples/GlassCard.tsx — a frosted card with aura.
  • assets/examples/RevealSection.tsx — scroll-triggered staggered reveal.
  • assets/examples/HeroBand.tsx — asymmetric two-column hero with page-load orchestration + scroll parallax.

Remember: match implementation effort to the vision. Maximalist directions need elaborate motion and layered effects; refined-minimal directions need restraint and obsessive spacing. Elegance is executing a clear vision fully — not turning every knob to 11.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Liquid Glass Frontend AI skill do?

Use when building a beautiful, distinctive, awwwards-tier web frontend — landing pages, dashboards, web apps, marketing sites — that needs Apple-style liquid-glass surfaces, choreographed motion.dev animations, and creative editorial layouts instead of the generic SaaS look (icon-card rows, centered heroes, default fonts, uniform fade-ins). Covers creative direction and signature interactions, the frost/blur glass material, single-source color tokens with dark+light theming, scroll/stagger/parallax/scroll-scrubbed animation, type-as-hero typography, and asymmetric grid-breaking composition.

Why use Liquid Glass Frontend on TypingMind?

Because you install it once and use it with any model. Liquid Glass Frontend is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Liquid Glass Frontend in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/ngocanhnckh/liquid-glass-frontend-skill/tree/main. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Liquid Glass Frontend?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Liquid Glass Frontend?

As many as you like. As long as a model supports skills, you can use Liquid Glass Frontend with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Liquid Glass Frontend AI skill free?

Yes. It is published on GitHub by ngocanhnckh under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇