PLAYWRIGHT
Identity: Browser automation and visual debugging tool Goal: Inspect pages, capture screenshots, debug UI issues
Usage
Use this skill when you need to:
- •Visual debugging of UI issues (like collections display)
- •Inspect DOM structure and console output
- •Capture screenshots of specific pages
- •Verify page state and interactions
Setup
Playwright CLI is already installed. Start a local server to test:
bash
cd /Users/liorshtram/dev/fermata/app && npm run dev
Common Commands
Navigation
bash
npx playwright open http://localhost:5173/collections
Screenshot (CLI)
bash
npx playwright screenshot http://localhost:5173/collections collections-view.png
Screenshot with full page
bash
npx playwright screenshot --full-page http://localhost:5173/collections collections-full.png
PDF export
bash
npx playwright pdf http://localhost:5173/collections collections.pdf
Interactive mode (headless: false)
bash
npx playwright open http://localhost:5173/collections --headless=false
Generate test from interaction
bash
npx playwright codegen http://localhost:5173/collections
Collections Debugging Workflow
- •Start the app:
cd app && npm run dev - •Open collections page:
npx playwright open http://localhost:5173/collections - •Capture initial state:
npx playwright screenshot http://localhost:5173/collections collections-initial.png - •Open DevTools in the browser to inspect elements
- •Capture after interaction:
npx playwright screenshot http://localhost:5173/collections collections-after.png - •Check console errors: Look at browser DevTools console
Integration with Tests
Run Playwright tests:
bash
cd /Users/liorshtram/dev/fermata/app npm run test:e2e
Notes
- •Use
headless: falseto see the browser and interact manually - •For CLI-only (agent workflow), use default headless mode
- •Screenshots saved to current directory by default
- •The app must be running for live URLs to work