Claude Code Plugins

Community-maintained marketplace

Feedback

Master prototyping - Figma, wireframes, interactive prototypes, user testing, iteration

Install Skill

1Download skill
2Enable skills in Claude

Open claude.ai/settings/capabilities and find the "Skills" section

3Upload to Claude

Click "Upload skill" and select the downloaded ZIP file

Note: Please verify skill by going through its instructions before using it.

SKILL.md

name prototyping
description Master prototyping - Figma, wireframes, interactive prototypes, user testing, iteration
sasmp_version 1.3.0
bonded_agent 04-prototyping
bond_type PRIMARY_BOND

Prototyping Skill

Atomic Skill: Create rapid, testable prototypes that validate design decisions

Purpose

This skill provides structured approaches to wireframing, prototyping, and design validation.

Skill Invocation

Skill("custom-plugin-ux-design:prototyping")

Parameter Schema

Input Parameters

interface PrototypingParams {
  // Required
  type: "wireframe" | "prototype" | "test" | "iterate";
  scope: string;

  // Optional
  fidelity?: "low" | "medium" | "high";
  platform?: "web" | "mobile" | "desktop";
  tool?: "figma" | "sketch" | "framer" | "html";
  testing?: {
    participants: number;
    scenarios: string[];
  };
}

Validation Rules

type:
  type: enum
  required: true
  values: [wireframe, prototype, test, iterate]

scope:
  type: string
  required: true
  min_length: 5

fidelity:
  type: enum
  default: "medium"

testing.participants:
  type: number
  min: 3
  max: 50
  default: 5

Execution Flow

PROTOTYPING EXECUTION
────────────────────────────────────────────

Step 1: DEFINE SCOPE
├── Identify key flows
├── Prioritize screens
└── Set fidelity level

Step 2: CREATE WIREFRAMES
├── Sketch layouts
├── Define content blocks
└── Map navigation

Step 3: BUILD PROTOTYPE
├── Apply visual design
├── Add interactions
└── Connect flows

Step 4: TEST WITH USERS
├── Prepare test script
├── Conduct sessions
└── Collect feedback

Step 5: ITERATE
├── Analyze findings
├── Prioritize changes
└── Update prototype

────────────────────────────────────────────

Retry Logic

retry_config:
  max_attempts: 3
  backoff_type: exponential
  initial_delay_ms: 1000
  max_delay_ms: 10000
  retryable_errors:
    - SYNC_FAILED
    - EXPORT_TIMEOUT
    - SESSION_INTERRUPTED

Logging Hooks

interface PrototypeLog {
  timestamp: string;
  event: "wireframe_created" | "prototype_built" | "test_started" | "iteration_complete";
  screens_count: number;
  interactions_count: number;
  test_sessions: number;
  success_rate: number;
}

Learning Modules

Module 1: Wireframing

WIREFRAME TYPES
├── Sketch (paper, whiteboard)
├── Low-fidelity (grayscale, boxes)
├── Mid-fidelity (structure, some detail)
└── High-fidelity (near-final)

WIREFRAME COMPONENTS
├── Layout grids
├── Content placeholders
├── Navigation elements
├── Form elements
└── Annotations

Module 2: Figma Workflows

FIGMA BEST PRACTICES
├── File organization
│   ├── Pages for flows
│   ├── Frames for screens
│   └── Components for reuse
├── Component structure
│   ├── Base components
│   ├── Variants
│   └── Instances
├── Prototyping
│   ├── Connections
│   ├── Interactions
│   └── Animations
└── Collaboration
    ├── Comments
    ├── Dev mode
    └── Handoff

Module 3: Interactive Prototyping

INTERACTION TYPES
├── Click/tap navigation
├── Hover states
├── Scroll behaviors
├── Input interactions
├── Animations
└── Smart animate

PROTOTYPE FLOWS
├── Happy path (success)
├── Error paths
├── Edge cases
└── Empty states

Module 4: User Testing

TEST PREPARATION
├── Define objectives
├── Write test script
├── Prepare scenarios
├── Set up recording
└── Recruit participants

TEST EXECUTION
├── Introduction (5 min)
├── Tasks (20-30 min)
├── Debrief (5 min)
└── Documentation

ANALYSIS
├── Task success rates
├── Time on task
├── Error frequency
├── Satisfaction ratings
└── Qualitative insights

Module 5: Iteration Cycles

ITERATION FRAMEWORK
├── Collect feedback
├── Identify patterns
├── Prioritize changes
├── Implement updates
├── Validate improvements
└── Document decisions

ITERATION VELOCITY
├── Quick wins (< 1 hour)
├── Standard changes (< 1 day)
├── Major revisions (< 1 week)
└── Structural changes (> 1 week)

Error Handling

Error Code Description Recovery
PT-001 Incomplete flow Add missing screens
PT-002 Broken link Fix connections
PT-003 Test invalid Refine methodology
PT-004 Scope exceeded Refocus on MVP
PT-005 Sync failed Manual backup

Troubleshooting

Problem: Users confused during testing

Diagnosis:
├── Check: Task clarity
├── Check: Prototype completeness
├── Check: Missing affordances
└── Solution: Improve setup

Steps:
1. Review task wording
2. Add interaction hints
3. Fill in missing screens
4. Brief users properly

Problem: Prototype doesn't match design

Diagnosis:
├── Check: Design system sync
├── Check: Component versions
├── Check: Manual overrides
└── Solution: Resync sources

Steps:
1. Update component library
2. Replace broken instances
3. Remove manual overrides
4. Verify with design team

Unit Test Templates

describe("PrototypingSkill", () => {
  describe("wireframe creation", () => {
    it("should include all specified screens", async () => {
      const result = await invoke({
        type: "wireframe",
        scope: "onboarding flow",
        screens: ["welcome", "signup", "profile", "complete"]
      });
      expect(result.screens.length).toBe(4);
    });
  });

  describe("prototype interactions", () => {
    it("should connect all screens", async () => {
      const result = await invoke({
        type: "prototype",
        scope: "checkout flow"
      });
      expect(result.orphaned_screens).toHaveLength(0);
    });
  });

  describe("test execution", () => {
    it("should generate success metrics", async () => {
      const result = await invoke({
        type: "test",
        testing: { participants: 5, scenarios: ["complete purchase"] }
      });
      expect(result.metrics.task_success_rate).toBeDefined();
    });
  });
});

Quality Metrics

Metric Target Measurement
Flow coverage > 90% Screens connected
Interaction completeness > 95% States defined
Test success rate > 80% Task completion
Iteration velocity < 24h Time per cycle

Version History

Version Date Changes
1.0.0 2025-12-30 Production-grade upgrade