Taste Skill

Kills the default LLM aesthetic before it ships. Reads your brief, infers the design direction from vibe words and audience signals, then sets three dials for variance, motion, and density before touching code. Maps you to real design systems when appropriate (Fluent, Carbon, GOV.UK) instead of recreating them badly. Built for landing pages, portfolios, and redesigns, not dashboards or data tables. The preflight discipline is the real value here: it forces a one-line design read and asks one clarifying question instead of dumping you into Inter plus purple gradients. Use this when you need a polished marketing site that doesn't look like every other Claude output.

install copies · 90 days. Times the install command was copied on this site in the last 90 days, bots filtered.: 400
skills.sh installs: 808
GitHub stars: 106.9K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add nexu-io/open-design --skill taste-skill --agent claude-code

Installs into .claude/skills of the current project.

Files

tasteskill: Anti-Slop Frontend Skill

Landing pages, portfolios, and redesigns. Not dashboards, not data tables, not multi-step product UI. Every rule below is contextual. None of it fires automatically. First read the brief, then pull only what fits.

0. BRIEF INFERENCE (Read the Room Before Anything Else)

Before touching code or tweaking dials, infer what the user actually wants. Most LLM design output is bad because the model jumps to a default aesthetic instead of reading the room.

SKILL.md on GitHub

Sponsors

Promote your product hereReach 125k+ visitors monthly Promote your product hereReach 125k+ visitors monthly