MCP HubMCP Hub
Volver a habilidades

tool-design-sprint-decide-and-storyboard

product-on-purpose
Actualizado 2 days ago
7 vistas
238
33
238
Ver en GitHub
Metadesign

Acerca de

Esta Habilidad de Claude facilita el día crítico de toma de decisiones (miércoles) de un Design Sprint, guiando a los equipos a través de métodos de votación estructurados para seleccionar el mejor boceto de solución. Luego, ayuda a traducir el concepto elegido en un storyboard detallado de 5 a 15 pasos que sirve como especificación para la construcción del prototipo. Úsala después de completar los bocetos para converger en un plan único y accionable para la creación del prototipo.

Instalación rápida

Claude Code

Recomendado
Principal
npx skills add product-on-purpose/pm-skills -a claude-code
Comando PluginAlternativo
/plugin add https://github.com/product-on-purpose/pm-skills
Git CloneAlternativo
git clone https://github.com/product-on-purpose/pm-skills.git ~/.claude/skills/tool-design-sprint-decide-and-storyboard

Copia y pega este comando en Claude Code para instalar esta habilidad

Documentación

<!-- PM-Skills | https://github.com/product-on-purpose/pm-skills | Apache 2.0 -->

Design Sprint Decide and Storyboard (Wednesday)

Wednesday is the most decision-heavy day of the Design Sprint. The skill runs the art museum layout (sketches posted anonymously on the wall), heat map (silent dot-vote stickers on promising parts), speed critique (3 minutes per sketch summarizing what the team sees), straw poll (each team member's top choice), and the Decider's supervote (the call that triggers the storyboard). Then the team decides whether to rumble (storyboard two sketches as competing prototypes) or all-in-one (storyboard one). The day ends with a 5-15 step storyboard that drives Thursday's build.

Family contract: docs/reference/skill-families/design-sprint-skills-contract.md. This skill is a member of design-sprint-skills.

When to Use

  • It is Day 3 of the Design Sprint and Tuesday's sketches are collected, attribution-stripped, and uploaded to the shared workspace.
  • The Decider is present for the full morning (heat map plus critique plus straw poll plus supervote happen 09:00-12:30 canonically).
  • The team has accepted that the supervote is the Decider's call, not a consensus average.
  • Thursday's prototype build needs a storyboard by Wednesday end-of-day or the build cannot start clean Thursday morning.

When NOT to Use

  • Tuesday is not closed. Return to tool-design-sprint-sketch; without independent sketches, the heat-map is meaningless.
  • Sketcher attribution leaked into Wednesday morning. The Facilitator must re-strip before the heat-map begins; voting on sketchers contaminates the day.
  • The Decider is absent. Wednesday's supervote cannot be delegated; without the Decider, the day produces a recommendation, not a decision.
  • The team is treating the supervote as advisory. The supervote is the call. Re-litigating it Thursday morning is sprint failure.

What This Skill Produces

A single bundled artifact with seven sections:

  1. Art museum layout: how sketches are arranged for viewing (typically wall or shared Figma board); attribution stripped; sketches labeled A / B / C / D / etc. in randomized order.
  2. Heat map: silent dot-vote stickers (typically 3 small dots per voter) placed on the most compelling PARTS of any sketch; voters can stack dots on the same element.
  3. Speed critique notes: 3-minute structured walkthrough per sketch by the Facilitator (or rotated team member) summarizing what the team sees; the sketcher does NOT explain their own sketch; concerns are captured as "what would worry me" notes.
  4. Straw poll results: each team member places one larger sticker (typically 1 large dot per voter) on the sketch they would pick if it were their call. Non-binding; informs the Decider.
  5. Supervote: the Decider places their supervote sticker (typically 3 large dots or a distinctively-colored sticker). The supervote is the decision. The Decider can place multiple supervotes on different sketches if running a rumble.
  6. Rumble vs all-in-one decision: the Decider chooses to storyboard one sketch (all-in-one; default) or two sketches as competing prototypes (rumble; chosen when the supervotes split clearly across two distinct approaches).
  7. Storyboard: 5-15 step storyboard derived from the supervoted sketch(es); each panel shows what the customer sees and what they do; specific enough that Thursday's build can begin without re-debating the design.

See references/TEMPLATE.md for the canonical structure and references/EXAMPLE.md for the Brainshelf book-catalog Wednesday artifact.

Wednesday Time Structure

The full Wednesday workshop is approximately 7 hours (09:00-12:30 + 13:30-17:00). The skill's bundled artifact emerges across the day:

  • 09:00-09:15: Welcome + Tuesday recap + Wednesday agenda
  • 09:15-09:30: Art museum setup; Facilitator confirms sketches are attribution-stripped and labeled A/B/C/D
  • 09:30-09:45: Heat map (silent; 15 minutes; team places dot stickers on compelling parts)
  • 09:45-10:45: Speed critique (3 min per sketch; sketchers silent during their own sketch's critique)
  • 10:45-11:00: Break
  • 11:00-11:15: Straw poll (silent; each person places 1 large dot on their pick)
  • 11:15-11:30: Decider's supervote and rumble vs all-in-one call
  • 11:30-12:30: Storyboard framing (which panels matter most; which map steps to cover; storyboard scope)
  • 12:30-13:30: Lunch
  • 13:30-16:30: Storyboard build (Facilitator drives; team contributes panel content; storyboard reaches 5-15 panels)
  • 16:30-17:00: Day-end review; Thursday build kickoff prep (who builds, who supports, what tools, what gets prebuilt overnight)

This skill's 210-minute timebox covers the Facilitator-led portions (heat map setup, critique, straw poll, supervote, storyboard build).

Common Pitfalls

  • Consensus drift instead of Decider supervote. The most common failure: the team votes, the Facilitator averages, and the Decider rubber-stamps the average. The Decider's call is the Decider's call; the straw poll is input, not result.
  • Storyboard too vague. "User captures a book" is not a storyboard panel; "user opens camera surface, points at book held at arm's length, sees recognized cover appear in bottom card within 1.5 seconds, taps the card to commit" is. Vague storyboards force Thursday's builders to re-debate design.
  • Skipping critique to save time. Critique surfaces concerns the storyboard must address. Skipping it produces a storyboard that hits Thursday and immediately re-surfaces every concern that should have been addressed Wednesday.
  • Sketcher explaining their own sketch. During critique, the sketcher is silent. The point is to test whether the sketch communicates without the artist. If the team can't read the sketch, that's data, not a bug.
  • Rumble chosen when all-in-one was right (or vice versa). Rumble is appropriate when 2 distinct supervoted approaches are both worth testing AND the team has Thursday capacity to build 2 prototypes. Default is all-in-one. Rumble is exception, not norm.
  • Storyboard scope too broad. 5-15 panels, not 50. The storyboard covers the target moment (from Monday) plus 1-2 panels of setup and 1-2 panels of outcome; it does NOT cover the full product.

Cross-Skill Usage

Prerequisites: tool-design-sprint-sketch. Decide and Storyboard consumes Tuesday's sketches as input. Without independent sketches, this skill has no material to heat-map.

This skill invokes tool-note-and-vote twice during the morning: once for the heat-map (small-dot voting) and once for the straw poll (large-dot voting). The supervote itself is the Decider's call after seeing both vote distributions; it does NOT delegate to note-and-vote.

Next invocation in the sprint: tool-design-sprint-prototype-plan Thursday morning.

Canonical Sources

Decider Checkpoint

This skill ends with a Decider Checkpoint in references/TEMPLATE.md. The Decider's supervote earlier in the day is the substantive decision; the end-of-day checkpoint confirms the storyboard is build-ready (specific enough; covers the target moment plus setup plus outcome panels; no remaining hand-waving). Without sign-off, Thursday's build begins with ambiguity and Friday's test risks being a test of the wrong artifact.

Repositorio GitHub

product-on-purpose/pm-skills
Ruta: skills/tool-design-sprint-decide-and-storyboard
0
agent-skillsai-skillsclaude-codeclaude-desktopdesign-sprintfoundation-sprint

Habilidades relacionadas

content-collections

Meta

Esta habilidad proporciona una configuración probada en producción para Content Collections, una herramienta centrada en TypeScript que transforma archivos Markdown/MDX en colecciones de datos con tipado seguro mediante validación Zod. Úsala al construir blogs, sitios de documentación o aplicaciones Vite + React con mucho contenido para garantizar seguridad de tipos y validación automática de contenido. Abarca todo, desde la configuración del plugin de Vite y compilación MDX hasta la optimización de despliegue y validación de esquemas.

Ver habilidad

polymarket

Meta

Esta habilidad permite a los desarrolladores crear aplicaciones con la plataforma de mercados de predicción Polymarket, incluyendo la integración de API para operaciones y datos de mercado. También proporciona transmisión de datos en tiempo real a través de WebSocket para monitorear operaciones en vivo y actividad del mercado. Úsela para implementar estrategias de trading o crear herramientas que procesen actualizaciones de mercado en tiempo real.

Ver habilidad

creating-opencode-plugins

Meta

Esta habilidad ayuda a los desarrolladores a crear complementos de OpenCode que se conectan a más de 25 tipos de eventos, como comandos, archivos y operaciones LSP. Proporciona la estructura del complemento, las especificaciones de la API de eventos y los patrones de implementación para módulos en JavaScript/TypeScript. Úsala cuando necesites interceptar, monitorear o extender el ciclo de vida del asistente de IA de OpenCode con lógica personalizada basada en eventos.

Ver habilidad

sglang

Meta

SGLang es un framework de alto rendimiento para el servicio de LLM que se especializa en generación rápida y estructurada para JSON, expresiones regulares y flujos de trabajo de agentes utilizando su caché de prefijos RadixAttention. Ofrece una inferencia significativamente más rápida, especialmente para tareas con prefijos repetidos, lo que lo hace ideal para salidas complejas y estructuradas, y conversaciones multiturno. Elige SGLang sobre alternativas como vLLM cuando necesites decodificación restringida o estés construyendo aplicaciones con uso extensivo de prefijos compartidos.

Ver habilidad