Skills for Real Engineers. Straight from my .agents directory.
39 skills
design-taste-frontend
Agent skill from
tasteskill 2.5.1, the anti-slop frontend skill. Builds professional company websites at the level of figma.com and x.ai: plain copy in a voice of its own, one system, one section per screen, equal edges, real photographs and real product UI (licensed stand-in photographs when the client has none; a product render only when the brief asks for 3D), a hero entrance and one or two dealt moments on every page, shader effects and scenes only on tech, gaming and software pages, quiet micro-interactions, and a checker and finish gate that enforce it in a real browser. Use this whenever you build, design, redesign, restyle, polish or review a landing page, company, product, shop, cafe, hotel, studio, SaaS, AI, developer-tool or game site, as static HTML/CSS or inside an existing React, Next.js, Vue, Svelte, Astro or Tailwind project, or when the user says a page looks generic, templated or "AI-made".
How to install design-taste-frontend
install design-taste-frontend
# install the taste-skill pack with the skills CLI
npx skills add Leonxlnx/taste-skill# or add only this skill to Claude Code: copy the folder
skills/taste-skill/
# into
~/.claude/skills/taste-skill/ (all projects)
.claude/skills/taste-skill/ (one project)
1# tasteskill 2.523Build the website a good company would ship: it says plainly what is sold, shows the real thing on4every screen, and looks made for this client, not from a template and not like an art piece.5figma.com and x.ai are the bar (in words: `layout-v4.md` section 1).67`<skill>` is this folder. It needs Node 22+ and Chrome, Edge or Chromium (ffmpeg optional).89## House rules1011These nine win over everything else in this skill.12131. **No caption or description under any image, render or visual.** Licence credits go in one14 footer line.152. **Every visual renders, never an empty ground,** at 1440, 1024, 768 and 390.163. **One section per screen:** the hero alone on the first screen, every section exactly one or17 two screens tall at 1440x900, no second section heading peeking in. The one exception is a18 live `photo-chapters` section, three or four screens, set by the moment.194. **Equal left and right content edges.**205. **Hover micro-animations:** a linked card lifts 2px, a photo zooms 1.02-1.03 inside its frame,21 buttons step on hover and press on click, underlines draw, card loops cycle in 2.5-4 s, repeat22 and restart on hover. The motion layer does it; never strip or override it. No tilt, carousels23 or timed tabs. Every hero enters once on load (`data-enter`), and every page carries one dealt24 moment, two on a long page (the gate holds both).256. **More and better real photographs where they fit;** nicer backgrounds; better heroes, layouts26 and fonts.27…
Excerpt shown under the repository's MIT license.
Keep exploring
Similar skill repositories
An agentic skills framework & software development methodology that works.
15 skills
A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.
CC1 skill
Public repository for Agent Skills
20 skills
Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.
CCCU6 skills
An AI skill that provides design intelligence for building professional UI/UX across multiple platforms.
CCCXCUGH6 skills