---
description: Execute comprehensive UI testing using Claude subprocess for complex flows
usage: /project:test-ui-subprocess [TEST-DESCRIPTION]
example: /project:test-ui-subprocess "Test the complete user registration and onboarding flow"
---
I'll execute comprehensive UI testing using a Claude subprocess for autonomous test execution.
## Step 1: Prepare Test Context
### Gather Information
!echo "Current branch: $(git branch --show-current)"
!echo "Application URL: [Will need to determine from config or ask user]"
### Verify Puppeteer MCP Availability
I'll launch a subprocess that can use the Puppeteer MCP tools for UI testing.
## Step 2: Determine Test Scope
Based on the feature and JIRA issue, I'll identify:
- User flows to test
- Critical UI interactions
- Visual validation requirements
- Edge cases to explore
## Step 3: Launch Subprocess Testing
I'll execute the UI tests using a Claude subprocess:
```bash
claude -p "You are testing a web application UI. Use the Puppeteer MCP tools to:
1. Navigate to [APPLICATION_URL]
2. Test the following user flow: [SPECIFIC_FLOW_DESCRIPTION]
3. Take screenshots at key points:
- Initial page load
- After each major interaction
- Final state
- Any error states encountered
4. Verify the following:
- All interactive elements work correctly
- Form validations trigger appropriately
- Navigation flows as expected
- Visual elements render correctly
- Responsive design works on different viewports
5. Test edge cases:
- Invalid inputs
- Rapid clicking
- Browser back/forward navigation
- Session timeout handling
6. Document your findings including:
- Steps performed
- Any errors or unexpected behaviors
- Performance observations
- Accessibility issues noticed
Use viewport sizes:
- Desktop: 1920x1080
- Tablet: 768x1024
- Mobile: 375x667
Start with headless: false to observe the test execution." --dangerously-skip-permissions
```
## Step 4: Process Results
After subprocess execution:
- Review the test output
- Analyze screenshots captured
- Identify any failures or issues
- Create summary of test results
## Step 5: Create Test Report
Generate a comprehensive test report including:
- Test scenarios executed
- Pass/fail status
- Screenshots of key states
- Performance metrics
- Accessibility findings
- Recommendations for fixes
## When to Use This Command
Use subprocess UI testing when:
- Testing complex multi-page workflows
- Need visual regression validation
- Exploring edge cases autonomously
- Testing across multiple viewport sizes
- Natural language test cases from JIRA
- End-to-end user journey validation
## Example Complex Flows
### E-commerce Checkout
```bash
"Test complete checkout: Browse products, add to cart, apply coupon, enter shipping, select payment, complete order, verify confirmation"
```
### User Onboarding
```bash
"Test new user flow: Sign up, verify email, complete profile, tutorial walkthrough, first action, dashboard familiarization"
```
### Admin Dashboard
```bash
"Test admin capabilities: Login as admin, manage users, configure settings, generate reports, handle notifications, test permissions"
```
---
## 🚀 Next Steps After Subprocess Testing
After the subprocess completes:
1. **Review the test output:**
- Check for any failures or unexpected behaviors
- Review screenshots for visual issues
- Note any performance concerns
2. **If tests pass, proceed to complete:**
```
/project:complete-issue [SITE-ALIAS]
```
3. **If issues found:**
- Fix the identified problems
- Run `/project:test-issue` for quick re-test
- Or run this command again for another deep exploration
💡 **Tip:** Subprocess testing excels at finding edge cases and visual issues you might miss!
