GetKits
KitsSkillsBotsDotsBundlesPreciosGuías
Kits/Diseño/Kit de estudio de diseño
BundleDiseñoOficial de GetKits

Kit de estudio de diseño

Kits de marca, críticas puntuadas y chequeos de lanzamiento para que la UI hecha con IA no se vea genérica.

por GetKits

Mira cómo las tres piezas trabajan juntas

El Skill hace el trabajo cuando se lo pides, el Bot ejecuta la rutina en Grok y el Dot vigila desde ChatGPT. Cada paso es un antes y después real de este Kit.

Ejemplo real · Kit de estudio de diseño

Antes

Nuestro hero, hecho en v0: degradado de violeta a azul, H1 centrado "Supercharge Your Workflow", subtítulo #94A3B8 sobre el degradado, dos botones tipo píldora, seis tarjetas de funciones con emojis debajo. ¿Esto parece hecho con IA?

Después

Puntuación: 15/35 Veredicto: RehacerSeñales genéricas detectadas: degradado violeta, todo centrado, sopa de tarjetas, iconos emoji, copy exagerado.1. Accesibilidad: subtítulo #94A3B8 sobre #7C3AED da 2.22:1. Corrección: hero sólido #FBFAF7, subtítulo #5A635E (5.95:1). Prioridad 92. Jerarquía: dos botones píldora iguales. Corrección: un botón bg-primary, el otro un enlace de texto. Prioridad 93. Color: elimina el degradado; un primario plano, solo para acciones. Prioridad 6

"¿Esto parece hecho con IA?", respondido con una puntuaciónAbrir el Skill

Mira la diferencia

De genérico de IA a acorde a la marca en tres semanas

Antes

Orbit, analítica de producto para equipos B2B. Nuestra página de precios hecha en Lovable se ve como la de cualquier otra startup de IA, y nuestro único diseñador trabaja a tiempo parcial. Queremos arreglarla y que siga arreglada.

Después

Página de precios: de 18/35 a 25/35 en dos semanas, de Rehacer a Corregir y lanzar. Semana 1: kit de marca Orbit v1.0 entregado como tokens de Tailwind; la primera revisión del viernes fijó la línea base. Semana 2: texto atenuado corregido (de 3.75 a 6.96:1) y un solo botón primario, ambas correcciones de Esfuerzo 1. Semana 3: el dot detectó una imagen de lanzamiento fuera de marca antes de que la redimensionaran. Deuda de diseño: 7 issues abiertos, 6 cerrados. Del brief a la página aprobada: 3 días.

Skill and Bot 5 points apart on one page

Antes

Thu Oct 22, Cursor: Leo asked the Skill to CRITIQUE a 1440 px screenshot of the new comparison page before pushing it to staging. 27/35, Fix then ship. At 5:00 PM the staging URL went on the Notion review list. Fri Oct 23, 3:00 PM, Bot weekly review: staging-preview.orbit.app/compare 22/35, Rework. #design, Leo: "same page, 5 points apart?? which one do we trust" You paste the Skill's CRITIQUE into the Bot: "Why are you 5 lower?"

Después

Bot, Fri Oct 23: the gap is viewport and source, not taste. Kit version: both used Orbit v1.0. Viewport: the Skill's header says "1440 px, Mobile not checked"; the Bot checked 1440 and 390 px. Source: the Skill scored a screenshot (confidence: medium); the Bot scored the live page and pressed Tab through it (confidence: high). | Dimension | Skill | Bot | Evidence only the live page or 390 px shows | | Accessibility | 4 | 2 | Plan toggle has no visible focus ring; at 390 px the 4-column table scrolls sideways with no cue | | Spacing rhythm | 4 | 3 | Sections keep 96 px padding at 390 px; the kit says 64 | | Typography | 4 | 3 | Table text drops to 11 px at 390 px | | Polish | 3 | 2 | Table header scrolls away after row 6; "Starter" is cut to "Sta..." at 390 px | Hierarchy, Color and Consistency: same scores from both. The Bot's 22/35 (Rework) is the score of record. Linear, after your DECISION (approve all): 1. Compare: Add a visible focus ring to the plan toggle (Urgent, Priority 9) 2. Compare: Make the first table column sticky below 768 px (High, Priority 6) 3. Compare: Set section padding to py-16 md:py-24 (High, Priority 6) Not drafted: 11 px table text, Priority 4 (I2 x R2 / E1), below the routine's Priority 6 line. Skill, in Cursor, Mon Oct 26: "Apply fixes 1 to 3 with our tokens, then CRITIQUE at 1440 and 390." ring-2 ring-ring ring-offset-2 on the toggle, sticky left-0 bg-card on the first column, py-16 md:py-24 on 6 sections; no new hex values. Self-CRITIQUE: 27/35, Fix then ship, both widths checked. Bot, Fri Oct 30: staging-preview.orbit.app/compare 26/35 (last week 22/35, +4), Fix then ship. Accessibility 2 to 4, Spacing rhythm 3 to 4, Polish 2 to 3. Skill and Bot now 1 point apart; all 3 Linear issues closed by merges.

Launch week with all three on duty

Antes

Orbit Insights launches Tue Nov 10. The dot watches Launches/Orbit Insights in Google Drive and #launch-insights in Slack. The landing page is on staging. Assets are arriving from two contractors.

Después

Mon Nov 2, Dot: hero-og.png Off-brand (headline in violet #8B5CF6, expected --primary #C2410C). You approved the drafted thread reply; the contractor re-exported it the same day. Wed Nov 4, Dot: staging-preview.orbit.app/insights, Verdict: Fix then ship (0 Off-brand, 3 Drift, 1 Nit). Card radius 16 vs 12, H2 at 700 vs 600, section padding 72 vs 96. You pasted the CRITIQUE into the Skill in Cursor: "Apply fixes 1 to 3 with our tokens." Result: 9 class changes, no new hex values. Fri Nov 6, Bot weekly review: orbit.app/insights 28/35, Fix then ship, no Off-brand items. 2 High issues drafted for after launch. Mon Nov 9, 4:00 PM, Dot pre-launch summary: 14 items, 12 On brand, 2 Approved by me, 0 Off-brand, 0 waiting on decisions. Tue Nov 10: launched on brand. Tue Nov 17: the dot's launch check-ins ended as scheduled.

Qué incluye

  • SkillSkill de estudio de diseñoTokens de marca, specs de página y críticas puntuadas para que la UI hecha con IA no se vea genérica.
  • BotBot de estudio de diseñoUn compañero en Grok que puntúa tus páginas publicadas cada viernes y redacta las correcciones.
  • DotDot de estudio de diseñoTu dot revisa cada recurso de lanzamiento contra tu kit de marca antes de que salga.
  • design-studio-kit-playbook.md

Sobre este Kit

Get a working design system practice in a week: a brand kit in your code, a scored review of every important page each Friday, and every launch asset checked against the kit before it ships. The Skill makes the design decisions inside the tool that writes your UI, the Grok Bot measures quality on a schedule, and your ChatGPT dot keeps launches on brand. All three share one brand kit, one rubric and one vocabulary, so a problem found by one is fixed by another without translation. Setup takes about 30 minutes across the three.

What you get

  • Design Studio Skill for Claude, ChatGPT, Codex and Cursor. Five modes (BRAND KIT, PAGE DESIGN, COMPONENT SPEC, CRITIQUE, HANDOFF), brand kits delivered as CSS variables and a Tailwind v4 theme, a critique rubric with scoring anchors, the 13 tells of generic AI UI, iOS and mobile rules, font pairings with licenses, computed contrast ratios and a worked example.
  • Design Studio Bot for Grok Bot. Five skills and three routines: a Friday review of your live and staging pages scored out of 35, brand audits against your Notion brand kit, a monthly competitor landing page scan, handoff notes, and Linear issue drafts that wait for your approval.
  • Design Studio Dot for your ChatGPT dot. An ongoing launch responsibility: it watches your Google Drive launch folder and Slack launch channel, checks every asset and page, prepares size and copy variants, and brings you the design calls.
  • The Design Studio Kit playbook, only in the bundle. The setup order (brand kit first), how the three hand work to each other, a weekly rhythm that takes about 30 minutes of your time outside launch weeks, the three metrics to track, a week-by-week first 30 days and an FAQ.
  • One shared contract. The same token names, CRITIQUE format and severity scale in every piece, one DECISION format for the Bot and Dot, plus a version line on the brand kit so you can see when any copy goes stale.

Who it is for

  • Founders and small product teams building with AI who want their product to look designed, and want proof each week that it is getting better.
  • Teams with one designer, or none, who need decisions written down once and then enforced.
  • Marketing and product teams running launches with several people producing assets.

What it will not do

  • Draw your logo or illustrations, or replace a designer's judgment on the brand's signature move.
  • Post, publish, file issues or edit shared files without your approval. Every outbound action is a DECISION you answer.
  • Sign in to private pages or read Figma files. It reviews what renders: live URLs, preview links and exported images.
  • Invent proof. Customer logos, metrics and testimonials stay [confirm] until you supply them.
  • Certify accessibility compliance. It flags the everyday problems; a formal audit is separate.

Bought separately

The Skill, the Bot and the Dot are $19 each, $57 for all three. The bundle is $45, saves you $12, and adds the Design Studio Kit playbook, which only the bundle includes. The playbook is what turns three tools into one practice: the order to set them up in, how a fix moves from the Bot's Friday review to the Skill in your code editor and back for a re-score, and the numbers that show whether design quality is improving.

Reseñas

Aún no hay reseñas.

$45

Incluido en Pro
Tiempo de configuración
~30 min
Plataformas
ClaudeChatGPTCodexCursorGrok
Se conecta con
SlackLinearNotionGoogle Drive
Revisado

Kit oficial de GetKits

Creado, probado y mantenido por el equipo de GetKits.

  • Creado y probado por GetKits
  • Incluye guía de instalación y actualizaciones
  • Cubierto por la política de reembolsos de GetKits

Más Bundles

BundleOficial
$39

Kit de creación de Bots

Convierte una tarea que haces a mano en un compañero de IA seguro y probado en menos de una hora.

GmailSlackNotionClaudeChatGPT
~45 min
BundleOficial
$45

Kit de fábrica de contenido

De una pieza larga a una semana de posts con tu voz, programados en Notion y con revisión semanal.

XNotionGoogle DriveClaudeChatGPT
~60 min
BundleOficial
$49

Kit de bucle externo de ingeniería

Gestiona el bucle externo de tu equipo: triaje diario, veredictos de CI, notas de versión y postmortems.

GitHubLinearSlackClaudeChatGPT
~50 min
GetKits

Descubre Skills, Bots, Dots y flujos de trabajo listos para usar en ChatGPT, Grok, Claude y más. Un Kit, una tarea, resultados desde el primer día.

Suscríbete a nuestro boletín

Marketplace

  • Explorar Kits
  • Skills
  • Bots
  • Dots
  • Bundles
  • GetKits Pro

Kits por área

  • Ventas
  • Reuniones
  • Contenido
  • Ingeniería
  • Diseño
  • SEO / GEO
  • Creación de Bots
  • Vida personal

Empresa

  • Sobre nosotros
  • Blog
  • Documentación
  • Contacto
  • Vende en GetKits

Legal

  • Política de privacidad
  • Términos del servicio
  • Política de reembolsos

© 2026 GetKits. Todos los derechos reservados.