KS
Killer-Skills

prototype-builder — Categories.community

v1.0.0
GitHub

About this Skill

Ideal for UI-focused AI Agents needing to create human-centric Storybook prototypes with explicit confidence levels and graceful failure handling. Elmer - AI Product Management Orchestrator

tylersahagun tylersahagun
[0]
[0]
Updated: 3/4/2026

Quality Score

Top 5%
46
Excellent
Based on code quality & docs
Installation
SYS Universal Install (Auto-Detect)
Cursor IDE Windsurf IDE VS Code IDE
> npx killer-skills add tylersahagun/elmer/prototype-builder

Agent Capability Analysis

The prototype-builder MCP Server by tylersahagun is an open-source Categories.community integration for Claude and other AI agents, enabling seamless task automation and capability expansion.

Ideal Agent Persona

Ideal for UI-focused AI Agents needing to create human-centric Storybook prototypes with explicit confidence levels and graceful failure handling.

Core Value

Empowers agents to design interactive user flows and implement AI feature states with transparency, showing receipts for every AI decision and making confidence levels explicit, leveraging Storybook stories and human-centric AI design standards.

Capabilities Granted for prototype-builder MCP Server

Building Storybook stories with trust-based AI features
Creating new UI prototypes with explicit confidence levels
Implementing AI feature states with graceful failure handling
Designing interactive user flows with transparent AI decision-making

! Prerequisites & Limits

  • Requires Storybook integration
  • Human-centric AI design standards must be followed
  • Explicit confidence levels and receipts for AI decisions are mandatory
Project
SKILL.md
5.1 KB
.cursorrules
1.2 KB
package.json
240 B
Ready
UTF-8

# Tags

[No tags]
SKILL.md
Readonly

Prototype Builder Skill

Specialized knowledge for building Storybook prototypes that meet human-centric AI design standards.

When to Use

  • Creating new UI prototypes
  • Building Storybook stories
  • Implementing AI feature states
  • Designing interactive user flows

Design Principles

Trust Before Automation

  • New features start as suggestions, not automations
  • Show receipts (evidence) for every AI decision
  • Make confidence levels explicit
  • Graceful failure > silent failure

Emotional Design

  • Visceral: Does it look trustworthy at first glance?
  • Behavioral: Does it work predictably?
  • Reflective: Does user feel augmented, not replaced?

Persona Awareness

  • Reps fear: Surveillance, replacement, embarrassment
  • Managers fear: Losing touch, surveillance culture
  • RevOps fear: Ungovernable AI, lack of auditability

Creative Exploration (Required)

For each major component, create 2-3 directions:

DirectionUser ControlTrust RequiredBest Persona
Option AMaximumLowNew users, skeptics
Option BBalancedMediumMost users
Option CMinimalHighPower users

Required AI States

Every AI feature needs ALL states:

StateVisualCopyAnimation
Loading (short)SpinnerNonePulse
Loading (long)Progress"Analyzing..."Transitions
SuccessCheck, mutedAffirmingScale+fade 150ms
ErrorWarningHonest + solutionGentle shake
Low ConfidenceMuted, dotted"I think..."None
EmptyIllustrationEncouragingFade in

Required: Flow Stories

Every prototype MUST include interactive journey stories:

typescript
1export const Flow_HappyPath: Story = { 2 render: () => <ComponentJourney />, 3}; 4 5export const Flow_ErrorRecovery: Story = { 6 render: () => <ComponentErrorJourney />, 7};
Prototype TypeRequired Flows
Simple feature1 (happy path)
AI feature2 (happy + error)
Complex workflow3+

Component Structure

Always use versioned folders in prototypes/src/components/:

prototypes/src/components/[Initiative]/
├── index.ts            # Re-exports latest
├── v1/
│   ├── Component.tsx
│   ├── Component.stories.tsx
│   ├── ComponentJourney.tsx
│   └── types.ts

Tech Stack

  • React 18 + TypeScript (strict)
  • Tailwind CSS
  • shadcn/ui from @/components/ui/

Storybook Title Pattern

typescript
1const meta = { 2 title: "Prototypes/[Initiative]/v1/[ComponentName]", 3 component: ComponentName, 4};

Chromatic Deployment

After building, deploy to Chromatic:

bash
1cd prototypes 2npm run build-storybook 3CHROMATIC_PROJECT_TOKEN="chpt_46b823319a0135f" npm run chromatic

Production URL: https://main--696c2c54e35ea5bca2a772d8.chromatic.com

Image Generation for Mockups (Cursor 2.4)

When a quick visual is needed before building full components, use image generation:

When to Generate Images

  • Early concept exploration - Before committing to code
  • Stakeholder communication - Quick visuals for feedback
  • Design exploration - Testing multiple directions rapidly
  • Architecture diagrams - Visualizing data flows or system design

How to Use

Describe the mockup in natural language:

"Generate an image of a dashboard showing user engagement metrics with a sidebar navigation, using a clean modern design with blue accent colors"

Best Practices

  • Reference existing design system colors/patterns in descriptions
  • Generated images save to assets/ by default
  • Use for exploration, not final implementation
  • Follow up with coded Storybook components for production

Example Prompts

Use CasePrompt Example
Dashboard mockup"A meeting insights dashboard with cards showing action items, sentiment trends, and speaker breakdown"
Mobile view"Mobile-first meeting summary view with collapsible sections and swipe actions"
Error state"Error message UI showing a failed CRM sync with retry option and helpful guidance"
Flow diagram"User flow diagram from meeting recording to CRM update showing each step"

Anti-Patterns

  • Single option - Always explore 2-3 directions
  • Missing states - All AI states required
  • States without flows - Always include Flow_* stories
  • Confident wrongness - Show uncertainty
  • Surveillance vibes - "Helps YOU" not "reports ON you"

Related Skills

Looking for an alternative to prototype-builder or building a Categories.community AI Agent? Explore these related open-source MCP Servers.

View All

widget-generator

Logo of f
f

widget-generator is an open-source AI agent skill for creating widget plugins that are injected into prompt feeds on prompts.chat. It supports two rendering modes: standard prompt widgets using default PromptCard styling and custom render widgets built as full React components.

149.6k
0
Design

chat-sdk

Logo of lobehub
lobehub

chat-sdk is a unified TypeScript SDK for building chat bots across multiple platforms, providing a single interface for deploying bot logic.

73.0k
0
Communication

zustand

Logo of lobehub
lobehub

The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboration, effortless agent team design, and introducing agents as the unit of work interaction.

72.8k
0
Communication

data-fetching

Logo of lobehub
lobehub

The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboration, effortless agent team design, and introducing agents as the unit of work interaction.

72.8k
0
Communication