Browsing With Playwright

This is full browser automation via Playwright MCP, not just fetching HTML. You get all the tools for real web interaction: navigate pages, click buttons, fill forms, take screenshots, extract data. The important bit is the shared browser context flag, which keeps state across multiple script calls so you're not starting fresh every time. Use it when you need to actually interact with a page like filling out multi-step forms or scraping dynamic content. For static pages, just use curl. The workflow docs are solid, showing form submission and data extraction patterns. One catch: you need to get element refs from snapshots before interacting, so there's always a two-step dance of snapshot then action.