CCM
/Skills
SkillsMCPMarketplacesDigestToolsAdvertise

This week in Claude

Every Monday: Claude Code, Agent SDK, MCP, and the Anthropic platform moves worth your time.

Skills by Category
Frontend DevelopmentBackend & APIsTesting & QASecurityDevOps & CI/CDGit & Pull RequestsDocumentationCode Review & QualityAI & Agent BuildingSkill Development
MCP Servers by Category
Sales & MarketingWeb & Browser AutomationDatabasesAI & LLM ToolsCloud & InfrastructureCommunication & MessagingDeveloper ToolsDesign & CreativeDocuments & KnowledgeSearch & Web Crawling
Marketplaces by Category
AI Agents & OrchestrationLLM IntegrationDevelopment ToolsFrontend & UIBackend & APIsDatabasesTesting & Code QualityDevOps & CloudSecurity & ComplianceGit & Version Control

Claude Code Marketplaces

Discover Claude Code plugins, extensions, and tools. Automatically updated directory of Anthropic Claude AI marketplaces with development tools, productivity plugins, and integrations.

Resources

  • Browse Skills
  • Browse MCP Servers
  • Browse Marketplaces
  • Plugins Reference

Community

  • About
  • Tools
  • Feedback
  • Privacy Policy
  • Advertise

Built for the Claude Code community with Claude Code by mertbuilds.com

Independent project, not affiliated with Anthropic

Frontend Design Ui Ux

ulpi-io/skills
1.7k installs3 stars
Summary

This is the skill you reach for when you need a proper design spec before building UI. It walks you through discovery, flow modeling, component specs, design tokens, and accessibility requirements, then packages everything for handoff to a frontend engineer or agent. It enforces the discipline of thinking through states, edge cases, and responsive behavior before writing code. The templates in references keep you honest about what "implementation ready" actually means. Use it when someone says "design this dashboard" or "spec the onboarding flow" and you need more than wireframes. It stops at the handoff and won't write production code, which is the right split if you want design decisions separated from implementation.

Install to Claude Code

npx -y skills add ulpi-io/skills --skill frontend-design-ui-ux --agent claude-code

Installs into .claude/skills of the current project.

CodeRabbit
CodeRabbit
AI writes the code. CodeRabbit catches the slop.
Try For Free →
Give your AI the whole web as clean markdownGive your AI the whole web as clean markdown
Give your AI the whole web as clean markdown
Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →
belt - the only tool your agent needs
belt - the only tool your agent needs
belt cli automatically finds the best tools and skills for your agent. image, video, music, tts...
one prompt install →
inference shell
inference shell
create and run specialised agents in minutes
build now →
MCP-ready Email SendingMCP-ready Email Sending
MCP-ready Email Sending
Plug Mailtrap into your AI workflow and let it handle the email.
Connect Mailtrap MCP →
Make your agent a DeFi expert
Make your agent a DeFi expert
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
Capacitor - Shared memory for your team’s coding agents.
Capacitor - Shared memory for your team’s coding agents.
Make coding agent sessions - Searchable, Shareable, Vendor-neutral & Scored.
Try For Free →
CodeScene MCP ServerCodeScene MCP Server
CodeScene MCP Server
Your agent targets a perfect 10 Code Health score. Deterministic. Every commit.
Try For Free →
CodeRabbit
CodeRabbit
AI writes the code. CodeRabbit catches the slop.
Try For Free →
Give your AI the whole web as clean markdownGive your AI the whole web as clean markdown
Give your AI the whole web as clean markdown
Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →
belt - the only tool your agent needs
belt - the only tool your agent needs
belt cli automatically finds the best tools and skills for your agent. image, video, music, tts...
one prompt install →
inference shell
inference shell
create and run specialised agents in minutes
build now →
MCP-ready Email SendingMCP-ready Email Sending
MCP-ready Email Sending
Plug Mailtrap into your AI workflow and let it handle the email.
Connect Mailtrap MCP →
Make your agent a DeFi expert
Make your agent a DeFi expert
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
Capacitor - Shared memory for your team’s coding agents.
Capacitor - Shared memory for your team’s coding agents.
Make coding agent sessions - Searchable, Shareable, Vendor-neutral & Scored.
Try For Free →
CodeScene MCP ServerCodeScene MCP Server
CodeScene MCP Server
Your agent targets a perfect 10 Code Health score. Deterministic. Every commit.
Try For Free →
Files
SKILL.mdView on GitHub
This is a DESIGN-SPEC skill, not an implementation skill. Non-negotiable rules: 1. Do NOT write production UI code from this skill. Produce a spec + a locked design language and DELEGATE the build to an engineering agent. We own the design spec; the agent implements it. 2. Commit to ONE distinctive aesthetic identity BEFORE specifying screens, and pass the counterfactual default test: if you'd produce this same look for any similar brief, it's a default — change it. 3. Ban AI-slop BY NAME (`references/anti-slop.md`). The slop test gates the work: if someone could say "AI made that" with no doubt, it failed. 4. LOCK the design language in `DESIGN.md` and bind every screen to it. Variation within identity, NEVER between. Re-read `DESIGN.md` first on every later feature or session. 5. If the user gives inspiration links, VISIT them with the `browse` skill and synthesize — never clone a single reference, and never inherit its slop. 6. Run the Design Pre-Flight gate before handoff. If a single box can't be honestly ticked, it is not done. 7. Keep the rigor that sets us apart: state coverage, flows/edge cases, and accessibility — never skip them.

frontend-design-ui-ux

Inputs

  • $request: Feature, page, component, or UX problem to design — plus optional inspiration URLs.

Goal

Produce a design package that is distinctive, consistent, and buildable — a spec a delegated engineering agent implements directly. All artifacts are written under .ulpi/design/:

  • .ulpi/design/DESIGN.md — the locked design language (identity, palette, type, scales, signature, voice, chosen design system). Project-wide; re-read first every session.
  • .ulpi/design/<feature>.md — the per-feature spec: user flows + states, component specs, and the build handoff (target agent + acceptance criteria). Binds to DESIGN.md.

Step 0: Discovery & inputs

Resolve: target users · product goal · device/usage context · existing design-system or brand constraints · accessibility expectations. Ask once: "Any reference sites or designs you admire? Paste 1–4 URLs."

If the problem is underspecified, ask — but at most a couple of focused questions; never a long interrogation. Success criteria: the problem is framed in user + product terms, and inspiration links (if any) are collected.

Step 1: Inspiration intake (optional)

If the user gave links, VISIT each with the browse skill, extract its design DNA (type, color, shape, spacing, layout), and distill a one-line DNA note per site. Synthesize across them — never clone one. If browse is unavailable or a link is unreachable, ask the user to describe the vibe and proceed.

Load references/inspiration-intake.md. Success criteria: each reference reduced to reusable DNA, with what to take and what to reject — or cleanly skipped.

Step 2: Design Read & aesthetic direction

Approach as a design lead who would never ship a templated look. Kill the default-aesthetic reflex before any tokens:

  • Emit a one-line Design Read (the feeling + the bet).
  • Commit to one named aesthetic direction from the menu in references/anti-slop.md; justify it from the brief (and the inspiration DNA, if any).
  • Run the counterfactual default test: would this look be your answer for ANY similar brief? If so, change it.
  • Optionally set taste dials (design_variance / motion_intensity / visual_density, 1–10) and match complexity to the direction (maximalist → elaborate; minimal → restraint, precision, whitespace).

Load references/anti-slop.md for the direction menu, the pattern vocabulary, and the named bans. Success criteria: a committed, brief-specific direction that is not a generic default.

Step 3: Derive & LOCK the design language → write .ulpi/design/DESIGN.md

Decide the register (brand vs product) and, for product, the design system to build on (see references/design-system-routing.md). Then derive and LOCK the identity: palette (OKLCH, tinted neutrals, 60-30-10, WCAG-checked), type roles paired on a contrast axis, ONE spacing/radius/motion scale, the Signature, and the voice. Write it to .ulpi/design/DESIGN.md — the consistency source of truth every screen binds to.

Load references/design-language-lock.md (artifact + lock rules), references/design-tokens-template.md (derivation method), and references/design-system-routing.md (brief→system map). Success criteria: a complete, locked .ulpi/design/DESIGN.md; no generic default identity.

Step 4: Model flows and states

Document the primary journey, branching/decisions, and the full state model — loading, empty, partial, success, error — plus edge cases (refresh, session expiry, back, offline). Write into .ulpi/design/<feature>.md. Use references/user-flow-template.md. Success criteria: behavior is covered across states, not just the happy path.

Step 5: Specify components and interaction rules

For each component define: purpose, variants, props/data, states, interaction feedback, responsive behavior, and accessibility (ARIA, keyboard, screen-reader, focus). Bind every visual value to .ulpi/design/DESIGN.md; append into the same .ulpi/design/<feature>.md. Use references/component-spec-template.md. Success criteria: a delegated engineer can build it without guessing, and nothing drifts from the locked identity.

Step 6: Design Pre-Flight gate

Run references/design-preflight.md end to end — identity lock, anti-slop, state coverage, accessibility, layout craft, cognitive load, and the scored self-critique. Several items are mechanical counts. Fix any failing box or any axis scored ≤ 2, state in one line WHAT you changed and WHY (revise-and-justify), then re-run. Success criteria: every box ticked and no critique axis ≤ 2.

Step 7: Build handoff (delegate the build)

Finalize the delegation brief at the end of .ulpi/design/<feature>.md: the target engineering agent, the chosen design_system (+ its setup note), and the acceptance criteria — with the instruction: "Implement exactly this spec. Theme the design system with our locked tokens; do NOT redesign or re-implement its components." The building is delegated to the agent; this skill never writes the UI.

Agent Selection

CriteriaTarget Agent
Server-side rendering / SEO / App Router / Server Actionsnextjs-senior-engineer
Pure SPA / client-only / CLI web UI / Electron-Taurireact-vite-tailwind-engineer
Static site with no SSREither (ask user)
UnclearAsk user

Success criteria: a delegated agent can pick up .ulpi/design/DESIGN.md + .ulpi/design/<feature>.md and implement it directly, without redesigning.

Guardrails

  • Do not implement production UI code from this skill; produce the spec and hand off.
  • Do not ship a generic default identity — derive palette and type per brief; the slop test gates output.
  • Do not let screens drift from DESIGN.md: variation within identity, never between.
  • Do not clone a single inspiration reference, and never inherit its slop.
  • Do not skip accessibility, state coverage, or responsive behavior — they are our edge.
  • Do not produce vague design prose when a concrete artifact is needed.
  • Do not block if browse or an image tool is absent — degrade gracefully and proceed from the brief.

When To Load References

  • references/inspiration-intake.md — visiting reference links with browse and extracting DNA (Step 1).
  • references/anti-slop.md — the direction menu, named AI-slop bans, and the slop test (Steps 2, 3, 6).
  • references/design-language-lock.md — the DESIGN.md artifact and the consistency lock rules (Step 3).
  • references/design-system-routing.md — the brief→design-system map + honesty rule (Step 3, product register).
  • references/design-tokens-template.md — the per-brief color/type derivation method + neutral scales (Step 3).
  • references/user-flow-template.md — user journeys and acceptance flows (Step 4).
  • references/component-spec-template.md — component briefs and interface contracts (Step 5).
  • references/design-preflight.md — the pre-handoff quality + consistency gate (Step 6).

Output Contract

Write under .ulpi/design/ and report:

  1. the Design Read + committed aesthetic direction (and inspiration DNA, if any)
  2. .ulpi/design/DESIGN.md — the locked design language (register, design system, palette, type, scales, signature, voice)
  3. .ulpi/design/<feature>.md — flows and states
  4. — component specs (same file)
  5. Pre-Flight result (every box ticked, no critique axis ≤ 2)
  6. build handoff — target agent + design system + acceptance criteria ("implement exactly this; do not redesign")
Featured
CodeRabbit
CodeRabbit
AI writes the code. CodeRabbit catches the slop.
Try For Free →
Give your AI the whole web as clean markdownGive your AI the whole web as clean markdown
Give your AI the whole web as clean markdown
Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →
belt - the only tool your agent needs
belt - the only tool your agent needs
belt cli automatically finds the best tools and skills for your agent. image, video, music, tts...
one prompt install →
inference shell
inference shell
create and run specialised agents in minutes
build now →
MCP-ready Email SendingMCP-ready Email Sending
MCP-ready Email Sending
Plug Mailtrap into your AI workflow and let it handle the email.
Connect Mailtrap MCP →
Make your agent a DeFi expert
Make your agent a DeFi expert
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
Capacitor - Shared memory for your team’s coding agents.
Capacitor - Shared memory for your team’s coding agents.
Make coding agent sessions - Searchable, Shareable, Vendor-neutral & Scored.
Try For Free →
CodeScene MCP ServerCodeScene MCP Server
CodeScene MCP Server
Your agent targets a perfect 10 Code Health score. Deterministic. Every commit.
Try For Free →
Categories
Frontend DevelopmentDesign & UI/UX
First SeenJun 3, 2026
View on GitHub

Recommended

More Frontend Development →
nextjs-react-redux-typescript-cursor-rules

mindrally/skills

nextjs react redux typescript cursor rules
654
186
tailwind-css-patterns

giuseppe-trisciuoglio/developer-kit

Utility-first CSS framework patterns for responsive, component-based styling with Tailwind v4.1+.
13.4k
304
syncfusion-react-dashboard-layout

syncfusion/react-ui-components-skills

syncfusion react dashboard layout
157
3
ui-component-patterns

supercent-io/skills-template

Modern React component patterns for building scalable, maintainable UI libraries.
10.7k
88
ui-ux-pro-max

binjuhor/shadcn-lar

Frontend UI/UX design intelligence - activate FIRST when user requests beautiful, stunning, gorgeous, or aesthetic interfaces. The primary skill for design decisions before implementation. 50 styles, 21 palettes, 50 font pairings, 20 charts, 8 stacks (React, Next.js, Vue, Svelte, SwiftUI, React Native, Flutter, Tailwind). Actions: plan, build, create, design, implement, review, fix, improve, optimize, enhance, refactor, check frontend UI/UX code. Projects: website, landing page, dashboard, admin panel, e-commerce, SaaS, portfolio, blog, mobile app, .html, .tsx, .vue, .svelte. Elements: button, modal, navbar, sidebar, card, table, form, chart. Styles: glassmorphism, claymorphism, minimalism, brutalism, neumorphism, bento grid, dark mode, responsive, skeuomorphism, flat design. Topics: color palette, accessibility, animation, layout, typography, font pairing, spacing, hover, shadow, gradient.
59
flutter-build-responsive-layout

flutter/skills

flutter build responsive layout
23.2k
2.6k