Design Research

Browser Tool Startup Sequence (for every teammate)

Every agent must follow this sequence when starting:

1. mcp__claude-in-chrome__tabs_context_mcp — get browser context
2. mcp__claude-in-chrome__tabs_create_mcp — create a new tab
3. mcp__claude-in-chrome__navigate — go to the target URL
4. mcp__claude-in-chrome__computer (action=screenshot) — capture visual state
5. mcp__claude-in-chrome__read_page — inspect DOM/accessibility tree
6. mcp__claude-in-chrome__javascript_tool — extract computed styles, class names, data
7. mcp__claude-in-chrome__find — locate specific elements by description
8. mcp__claude-in-chrome__computer (action=scroll) — see content below the fold

Key tools by purpose

Research Areas (10 Tasks)

Task 1: Site Structure & Navigation

• Map all top-level URLs and page types (home, product, content, settings, etc.)
• Document the navigation hierarchy (breadcrumbs, sidebar, tabs, footer links)
• Identify navigation redundancy patterns (how many ways to get to the same content)
• Note URL patterns and routing conventions
• Identify navigation gaps (missing back-links, dead ends, no search, etc.)

Task 2: Page Layout & Chrome (per page type)

• For each distinct page type, document the shared "chrome" (header, sidebar, footer)
• Measure column layout (widths, sticky behavior, responsive breakpoints)
• Document the top bar elements (icons, toggles, user menu)
• Note which chrome elements appear/disappear per page type (build a Page Chrome Matrix)
• Record CSS classes and measurements for key layout containers

Task 3: Design System — Colors & Typography

• Extract exact color values (RGB/hex) for backgrounds, text, links, borders, accents
• Document the font stack (families, sizes, weights, line-heights per element)
• Capture syntax highlighting color scheme if code is present
• Note the color hierarchy and how it creates depth/layering
• Document the button system (variants, sizes, states)
• Identify whether CSS custom properties or utility classes are used

Task 4: Interactive Components Inventory

• Catalog every interactive component: forms, modals, accordions, toggles, collapsibles, carousels, players, search, etc.
• Document each component's states (collapsed/expanded, active/inactive, loading, error, disabled)
• Note toolbar buttons and their functions
• Identify the underlying libraries (if visible in DOM/network)

Task 5: Content Type Templates (per type)

• For each distinct content/page type, document:
  • The exact content flow (e.g., H1 → hero → body → sidebar → CTA → footer)
  • Unique elements per type
  • Variants within the type
• Build a Content Template Matrix showing which features appear on which page types

Task 6: UX Patterns & Micro-interactions

• Progressive disclosure patterns (what's hidden by default, how it reveals)
• State persistence (localStorage, cookies, URL params, session tracking)
• Gamification elements (progress bars, badges, streaks, points, rewards)
• Callout/info box styles and when each is used (info vs warning vs success vs error)
• Hover effects, transitions, and animations
• Loading states and skeleton screens

Task 7: Visual Design & Personality (qualitative)

• Illustrations, icons, and visual metaphors
• Whimsical or distinctive details (dividers, badges, mascots, copy tone)
• Typography warmth and font pairing analysis
• Overall emotional tone and brand personality
• Comparison to common design paradigms (minimal, editorial, playful, corporate, etc.)
• What makes this design memorable or distinctive

Task 8: Mobile & Responsive Behavior

• Resize to mobile breakpoint and document what changes
• Sidebar → drawer conversions
• Touch targets and hamburger menu behavior
• Features that are desktop-only
• Critical mobile issues (missing navigation, undersized targets, viewport bugs)
• Document all breakpoints used

Task 9: Technical Implementation

• Identify the framework (React, Vue, Next.js, Nuxt, Svelte, etc.)
• Identify CSS approach (Tailwind, styled-components, CSS modules, vanilla, etc.)
• Key component/class naming conventions visible in DOM
• Libraries and services (video hosting, analytics, auth providers, CDNs, etc.)
• API endpoints and data fetching patterns visible in network tab
• SEO implementation (meta tags, OG tags, structured data, canonical URLs)

Task 10: Accessibility & Semantics

• Heading hierarchy (proper H1 → H2 → H3 nesting)
• ARIA attributes on interactive elements
• Semantic HTML usage (main, nav, header, footer, aside, section, article)
• Keyboard navigability of key components
• Skip-to-content links
• Color contrast adequacy (check muted text on light backgrounds especially)
• Screen reader friendliness (labeled inputs, alt text, etc.)
• Assign an overall accessibility grade (A through F)

Execution Strategy

Dependencies

Three-Pass Execution

First pass (5 agents, parallel, no dependencies): Tasks 1, 3, 7, 9, 10

Second pass (4 agents, parallel, after Task 1 completes): Tasks 2, 4, 5, 6

Third pass (1 agent, after Task 2 completes): Task 8

Team Orchestration Procedure

Follow these steps exactly:

1. Create the team: TeamCreate with name like design-research-{site}
2. Create all 10 tasks: Use TaskCreate for each research area
3. Set dependencies: Use TaskUpdate with addBlockedBy to wire up the dependency chain
4. Spawn first-pass agents (5 agents for Tasks 1, 3, 7, 9, 10):
   • Each agent prompt must include:
     • The target URL
     • The specific research task description (copy from above)
     • The browser tool startup sequence
     • Instruction to report findings back via SendMessage
   • Use team_name parameter on the Task tool to attach agents to the team
5. As agents complete:
   • Mark their tasks as completed via TaskUpdate
   • Send shutdown requests via SendMessage
   • Approve shutdown responses
6. When Task 1 completes: spawn second-pass agents (Tasks 2, 4, 5, 6), passing the site map from Task 1's findings
7. When Task 2 completes: spawn third-pass agent (Task 8), passing the layout findings
8. After all agents complete — SYNTHESIZE: This is your most important job as lead. Read through ALL agent reports and synthesize them into a single cohesive markdown document at the output path the user specified (default: notes/design-research.md). Do not just paste reports together — resolve duplicates, ensure consistent formatting, add cross-references between sections (e.g., link color findings to accessibility contrast issues), and produce a polished, well-organized reference document.
9. Clean up: TeamDelete to remove team resources

Output Format

Compile all findings into a single markdown file with this structure:

# {Site Name} — Design & UX Research Report

> **URL**: {target URL}
> **Date**: {today's date}
> **Summary**: One-line description of the site's design approach

---

## Table of Contents
[Auto-generated from sections below]

## 1. Site Structure & Navigation
[Findings from Task 1]

## 2. Page Layouts & Chrome
[Findings from Task 2]

## 3. Design System — Colors & Typography
[Findings from Task 3 — include color tables with hex values and Tailwind equivalents]

## 4. Interactive Components
[Findings from Task 4]

## 5. Content Type Templates
[Findings from Task 5]

## 6. UX Patterns & Micro-interactions
[Findings from Task 6]

## 7. Visual Design & Personality
[Findings from Task 7]

## 8. Mobile & Responsive Behavior
[Findings from Task 8]

## 9. Technical Implementation
[Findings from Task 9 — include architecture diagram]

## 10. Accessibility & Semantics
[Findings from Task 10 — include grade and top recommendations]

Use tables, code blocks, and structured formatting throughout. Include exact CSS values, class names, hex colors, and measurements — not vague descriptions.