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 ngocanhnckh/liquid-glass-frontend-skill --skill liquid-glass-frontend-skillgit clone --depth 1 https://github.com/ngocanhnckh/liquid-glass-frontend-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/ngocanhnckh/liquid-glass-frontend-skill/liquid-glass-frontend-skill)<a href="https://agentmods.dev/skills/ngocanhnckh/liquid-glass-frontend-skill/liquid-glass-frontend-skill"><img src="https://agentmods.dev/badge/skills/ngocanhnckh/liquid-glass-frontend-skill/liquid-glass-frontend-skill/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/ngocanhnckh/liquid-glass-frontend-skill/liquid-glass-frontend-skill"><img src="https://agentmods.dev/badge/skills/ngocanhnckh/liquid-glass-frontend-skill/liquid-glass-frontend-skill.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 | $0.00124 | $0.02057 |
| Opus 5.5 | $0.00050 | $0.00823 |
| Sonnet 5.5 | $0.00025 | $0.00411 |
| Haiku 4.5 | $0.00012 | $0.00206 |
Grade C, and why
liquid-glass-frontend scanned grade C with 1 finding 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 2mo 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
- **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. How it starts
The opening of the file, as written. The whole thing — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.
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:
- 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. - 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.
- 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. - 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. - 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.mdandreferences/creative-direction.md. - Liquid glass means real frost, not transparency. A see-through
rgba()box is NOT glass. Glass = heavybackdrop-filter: blur()+saturate()+ a specular top edge + a colored aura behind it so the frost has something to refract. Seereferences/liquid-glass.md. - One source of truth for color. The entire palette derives from three RGB triplets. Re-skinning is editing three lines. See
references/theming.md.
What ships with it
10 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.
- assets/examples/GlassCard.tsx 1.2 KB
- assets/examples/HeroBand.tsx 3.3 KB
- assets/examples/RevealSection.tsx 1.3 KB
- assets/globals.css 11 KB
- references/creative-direction.md 6.3 KB
- references/layout-composition.md 6.8 KB
- references/liquid-glass.md 7.0 KB
- references/motion.md 5.9 KB
- references/theming.md 4.3 KB
- references/typography-color.md 4.6 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.
- 2mo ago First seen · 80 lines · 124 tokens per session scan C d73de255bea4
liquid-glass-frontend is a skill published in the GitHub repository ngocanhnckh/liquid-glass-frontend-skill (11 stars, last pushed 4mo ago), licensed MIT. It adds 124 tokens to every session and 2,057 once invoked, about $0.0005 per session on Opus 5.5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
liquid-glass-frontend compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| liquid-glass-frontendngocanhnckh/liquid-glass-frontend-skill | Skill | 11 | 4mo ago | MIT | C | 124 | Claude Code, Codex |
| liquid-glassstormaref/LiquidGlassSkill | Skill | 18 | 2mo ago | MIT | A | 69 | Claude Code |
| huashu-designalchaincyf/huashu-design | Skill | 29k | 2d ago | MIT | A | 123 | Claude Code, Codex |
| kiranism-shadcn-dashboardKiranism/next-shadcn-dashboard-starter | Skill | 7.2k | 3d ago | MIT | A | 142 | Claude Code, Codex |
| image-to-codeplugin87/ux-ui-agent-skills | Skill | 3.1k | 10d ago | MIT | A | 77 | Claude Code |
| animation-patternsvibeeval/vibecosystem | Skill | 539 | 2d ago | MIT | A | 26 | Claude Code, Codex |
| shiny-bslib-themingposit-dev/skills | Skill | 554 | 4d ago | MIT | A | 87 | Claude Code |