Skill Frontend Excellence logo

Skill Frontend Excellence

Organization
IAPro-Community
skill-frontend-excellence

Use for product frontend work: classify intent, discover the project's design system and Design Profile, preserve or evolve UI, run Visual QA, and refuse to mark frontend done from build-only evidence.

Overview

PublisherIAPro-Community
RepositoryOrquestrador-Maestro
Skill nameskill-frontend-excellence
Stars
63
Forks
11
Bundled files
22
LicenseApache-2.0
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.

  • 22 bundled files

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

  • Open source

    Published by IAPro-Community on GitHub. Read the source before you install it.

Installation

Install the Skill Frontend Excellence 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/IAPro-Community/Orquestrador-Maestro/tree/v1/codex/skills/skill-frontend-excellence
  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/IAPro-Community/Orquestrador-Maestro.git /tmp/Orquestrador-Maestro
mkdir -p .claude/skills
cp -r /tmp/Orquestrador-Maestro/codex/skills/skill-frontend-excellence .claude/skills/skill-frontend-excellence
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Skill Frontend Excellence 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 Skill Frontend Excellence 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 Skill Frontend Excellence 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.

Frontend Excellence

Use this as the process layer for product UI. It does not own design-system tokens or application brand.

text
Design system  = source of truth (components, tokens, themes, patterns)
This skill     = process (intent, creativity, preserve vs redesign, DoD)
Application    = product context (configured Design Profile)
Visual QA      = smoke evidence
Maestro        = orchestration / retries
Router         = discovery / invocation

Do not copy design-system documentation into this skill. Locate live metadata via adapters/design-system.md or the project's configured adapter.

This skill is for product surfaces (admin, authentication, portals, forms, tables). For cinematic marketing sites use skill-premium-web-experience. For a single visual polish pass use skill-impeccable. This skill coordinates those instead of replacing them.

Composition and precedence

Use one owner for each concern:

ConcernPrimary ownerDelegation rule
Intent, posture, Design Profile, and Definition of Doneskill-frontend-excellenceAlways load for product-frontend work.
Product-wide IA, navigation, screen inventory, and journeysskill-product-ux-architectureLoad when the problem spans multiple screens or the whole app.
Brand direction or a new visual languageskill-open-design-uiLoad only for explicit exploration or a requested visual system.
High-fidelity implementation craftskill-design-engineering-craftLoad after direction is clear when optical alignment, density, typography rhythm, and finish are the problem.
Motion-system design or animation reviewskill-motion-design-principlesLoad only when motion itself is material to the task.
Component states, interaction patterns, and polishskill-modern-ui-patternsLoad when the task changes interaction behavior or state coverage.
Responsive/accessibility usability gateskill-frontend-ux-guardrailsLoad for a UX gate or when responsive/accessibility risk is material.
Browser flow and E2E evidenceskill-webapp-testingLoad for a critical user journey or browser regression.
One-off conversational visual polishskill-impeccableUse instead of open-design for a bounded polish pass.
Cinematic marketing experienceskill-premium-web-experienceIt owns the strategy for marketing surfaces; do not also treat this skill as the top-level owner.

When more than one downstream skill applies, load at most two for a standard task. The top-level intent and Definition of Done remain owned here.

When to load

Natural-language requests such as:

  • modernize esta tela
  • migre MUI, Bootstrap ou outro design system para o sistema adotado pelo projeto
  • melhore este formulário
  • corrija o mobile
  • refatore este frontend
  • crie uma tela nova

Also load when the user asks for Visual QA, Design Profile, or frontend Definition of Done.

Step 0 — Classify intent (mandatory)

Derive, do not ask unless two intents are equally likely:

IntentTypical verbs
CREATEcrie, adicione, nova tela, novo fluxo
REDESIGNmodernize, redesenhe, nova identidade
IMPROVEmelhore, refine, acessibilidade, mobile
MIGRATEmigre MUI/Bootstrap ou outro design system para o sistema adotado pelo projeto
REFACTORrefatore, TypeScript, router, performance
FIXcorrija, bug, overflow, contraste
REVIEWrevise, audite, critique

Posture:

PostureMeaning
PRESERVEVisual change not requested is a regression
EVOLVELocal UX change inside existing identity
EXPLORENew composition is allowed after directions

Creativity (NONE|LOW|MEDIUM|HIGH) comes from the request and the application Design Profile. The stricter of the two wins.

Gold rule: refactor is not redesign. MIGRATE, REFACTOR, library upgrades, and React/router/TypeScript work default to PRESERVE + NONE.

Run scripts/classify-intent.mjs when available. Read only the matching workflow:

Step 1 — Resolve the design system (mandatory)

Run node scripts/discover-design-system.mjs <project-root> and persist its structured result. Only status: resolved with confidence >= 0.80 permits new visual implementation. For ambiguous or unresolved, stop with BLOCKED_REQUIRES_USER_DECISION; ask whether to preserve local components, use a named project system, or explicitly propose a new visual base. Never choose by package popularity or model preference.

Step 2 — Load context

  1. Application Design Profile: use the project's configured file, package export, or the bundled neutral fallback. Do not assume a repository layout.
  2. Design-system metadata via the adapter. Never invent components.
  3. Existing routes, tokens, i18n, tests, and startup commands from the app.
  4. If PRESERVE: capture baseline screenshots before editing.

Step 3 — Implement under the right workflow

  • CREATE → workflows/create.md
  • REDESIGN + HIGH → workflows/redesign.md (directions before code)
  • IMPROVE → workflows/improve.md
  • MIGRATE / REFACTOR + PRESERVE → workflows/migrate.md
  • FIX → workflows/fix.md
  • REVIEW → workflows/review.md

Engineering, visual, accessibility, responsive, and maintainability rules are in standards.

Step 4 — Visual QA is part of the work

Build passing is not done. Run the checklist in validation/definition-of-done.md, with smoke details in validation/visual.md and validation/accessibility.md.

Preferred command (the path is relative to this skill package):

bash
node scripts/visual-qa.mjs --url <app-url> --out .frontend-excellence/visual-qa/

Viewports default to the shared smoke widths: 390, 768, 1280, 1440.

Hard failures (any one fails the task):

  • console / page errors
  • contrast below WCAG AA for normal text (4.5:1) except a documented design-system exception
  • missing accessible name on controls
  • page-level overflow on required viewports
  • PRESERVE: unintended visual delta vs baseline

This is a smoke harness, not a complete WCAG audit. Score (validation/visual.md) never overrides a hard failure; use the project's accessibility tooling for certification.

Step 5 — Independent visual review

Produce structured findings even if the same agent implements:

yaml
id: UI-001
severity: high
viewport: 390x844
component: LoginPanel
problem: Insufficient text contrast.
expected: WCAG AA 4.5:1
actual: measured or approximated ratio
action: Use a semantic text token that meets AA.

Do not conclude with “ficou bonito” or “ficou profissional”.

Guardrails

  • Never force every app to look the same. A design system is vocabulary; the Design Profile is identity.
  • Never recreate a design-system component. Search metadata first.
  • Never start REDESIGN+HIGH by writing components. Produce 2–3 genuinely different directions first.
  • Never mark frontend complete from compile/lint alone.
  • Never overwrite uncommitted human work.
  • Prefer the project's documented package subpath imports when the app already does; do not introduce a barrel that the tree-shaking contract forbids.
  • Delegate visual craft to skill-open-design-ui, interaction states to skill-modern-ui-patterns, UX gate to skill-frontend-ux-guardrails, E2E to skill-webapp-testing.

Related Skills

  • skill-open-design-ui
  • skill-modern-ui-patterns
  • skill-frontend-ux-guardrails
  • skill-webapp-testing
  • skill-impeccable
  • skill-premium-web-experience
  • skill-verification-before-completion
  • skill-browser-agent

Verification

  • Intent, posture, and creativity are written down before the first UI edit.
  • Design-system metadata was consulted (or the adapter recorded that no supported system was found).
  • Design Profile was loaded or explicitly recorded as missing.
  • Visual QA artifacts exist under the project's configured output directory or the skill fixture run.
  • Definition of Done checklist is complete for the intent.

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 Skill Frontend Excellence AI skill do?

Use for product frontend work: classify intent, discover the project's design system and Design Profile, preserve or evolve UI, run Visual QA, and refuse to mark frontend done from build-only evidence.

Why use Skill Frontend Excellence on TypingMind?

Because you install it once and use it with any model. Skill Frontend Excellence 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 Skill Frontend Excellence in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/IAPro-Community/Orquestrador-Maestro/tree/v1/codex/skills/skill-frontend-excellence. 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 Skill Frontend Excellence?

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 Skill Frontend Excellence?

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

Is the Skill Frontend Excellence AI skill free?

Yes. It is published on GitHub by IAPro-Community under the Apache-2.0 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 👇