browsing-with-playwright

Automate browser interactions via Playwright MCP server. Important: The --shared-browser-context flag is required to maintain browser state across multiple mcp-client.py calls. Without it, each call gets a fresh browser context. Use ref from snapshot output to target elements: For complex workflows, use browser_run_code to run multiple actions in one call: Tip: Use browser_run_code for complex multi-step operations that should be atomic (all-or-nothing). Run: python3 scripts/verify.py Expected: ✓ Playwright MCP server running See references/playwright-tools.md for complete tool documentation. - End of task : Stop when browser work is complete - Long sessions : Keep running if doing multiple browser tasks - Errors : Stop and restart if browser becomes unresponsive - Navigate to page - Get snapshot to find element refs - Fill form fields using refs - Click submit - Wait for confirmation - Screenshot result - Navigate to page - Get snapshot (contains text content) - Use browser_evaluate for complex extraction - Process results - Run diagnostic: pgrep -f "@playwright/mcp" - Check: Server process running on port 8808 - Try: bash scripts/start-server.sh - Stop and report if still failing - do not proceed with downstream steps