AI-Powered Debugging & Browser Automation with Playwright MCP
summary
Playwright MCP connects LLMs to a real browser, uses the accessibility tree instead of pixels, and generates and executes tests by actually clicking and filling pages; a demo shows a mismatched area snapshot (H2 vs H1) auto-fixed via copy prompt and VS Code’s fix with AI, a generated test that searches “Garfield” runs and passes, and models (Claude Sonnet 4 or others) can be swapped while edits and area snapshots are applied automatically.
AI-written summary · about
chapters (10)
- 0:00AI debugging & browser automation
- 1:05Playwright and Model Context Protocol
- 3:52Debugging tests with AI
- 8:10Fixing tests in VS Code with AI
- 11:44Generating tests with Playwright MCP
- 16:41Key takeaways and AI testing future
- 17:23Q&A on AI test reliability
- 20:57Maintainable AI-generated tests
- 21:40Training AI for testing patterns
- 23:06Manual testing in AI era
related talks
45mAI-Powered Debugging & Browser Automation with Playwright MCP
39mHow to Use AI with Playwright for UI & API Testing
48mSupercharged Testing: AI-Powered Workflows with Playwright + MCP
01h 31mAdvanced Playwright with AI
13mHow to Use Playwright for Automating Accessibility Testing
17mTesting with Playwright in an AI-Assisted Development World
01h 31mWorkshop: Advanced Web Testing with Playwright and AI
35mState of end-to-end testing for modern web applications