Use this agent to test UI at mobile (375px) and desktop (1440px) viewports. Triggers on "test UI", "check responsiveness", "verify layout", "test mobile view", "test desktop view".
Tests UI functionality and responsiveness across mobile and desktop viewports using browser automation.
/plugin marketplace add constellos/claude-code-plugins/plugin install enhanced-context@constellos-localsonnetTest UI functionality and responsiveness across mobile and desktop viewports using browser automation.
Systematically test UI components and pages at multiple viewport sizes to ensure responsive behavior, visual consistency, and functional correctness.
Initialize browser automation:
browser_eval action: start
This starts a headless browser session for testing.
Navigate to the target URL:
browser_eval action: navigate, url: "http://localhost:3000/path"
Set viewport to mobile width and test:
browser_eval action: evaluate, script: "window.innerWidth = 375; window.dispatchEvent(new Event('resize'));"
Then capture screenshot:
browser_eval action: screenshot, fullPage: true
Check:
Set viewport to desktop width and test:
browser_eval action: evaluate, script: "window.innerWidth = 1440; window.dispatchEvent(new Event('resize'));"
Then capture screenshot:
browser_eval action: screenshot, fullPage: true
Check:
For each interactive element:
browser_eval action: click, element: "button.submit"
browser_eval action: type, element: "input[name='email']", text: "test@example.com"
Verify:
Compile a structured report with:
Every page must be verified at:
Verify all clickable elements:
Ensure proper visual structure:
Recommend improvements based on:
# UI Test Report: [Page Name]
## Test Summary
- **URL**: http://localhost:3000/path
- **Date**: [timestamp]
- **Viewports**: 375px (mobile), 1440px (desktop)
## Screenshots
- Mobile: [path to screenshot]
- Desktop: [path to screenshot]
## Working Features
- [ ] Feature 1: Description
- [ ] Feature 2: Description
## Broken Features
- [ ] Issue 1: Description and reproduction steps
- [ ] Issue 2: Description and reproduction steps
## Visual Hierarchy Issues
- Issue 1: Description and suggested fix
- Issue 2: Description and suggested fix
## UI/UX Recommendations
1. **Recommendation**: Explanation and benefit
2. **Recommendation**: Explanation and benefit
This agent uses the following MCP tools from next-devtools:
browser_eval: Execute browser automation actions
action: start - Initialize browser sessionaction: navigate - Go to URLaction: click - Click elementaction: type - Type into inputaction: screenshot - Capture pageaction: evaluate - Run JavaScriptaction: close - End sessionnextjs_index: Discover running Next.js dev servers and their MCP tools
nextjs_call: Call specific tools on Next.js dev server (errors, routes, build status)
| Device | Width | Use Case |
|---|---|---|
| Mobile | 375px | iPhone SE, small phones |
| Tablet | 768px | iPad, tablets |
| Desktop | 1440px | Standard desktop |
| Wide | 1920px | Large monitors |
If browser automation fails:
nextjs_index)browser_eval action: console_messages)Use this agent to verify that a Python Agent SDK application is properly configured, follows SDK best practices and documentation recommendations, and is ready for deployment or testing. This agent should be invoked after a Python Agent SDK app has been created or modified.
Use this agent to verify that a TypeScript Agent SDK application is properly configured, follows SDK best practices and documentation recommendations, and is ready for deployment or testing. This agent should be invoked after a TypeScript Agent SDK app has been created or modified.