Browser rendering

The use_browser option controls how BeeCrawl fetches a page:

Value	Behavior
`never`	Use plain HTTP only.
`always`	Use Bee Engine and Playwright only.
`auto`	Try Playwright first, then fall back to HTTP when rendering fails or produces no Markdown.

{
  "url": "https://example.com/app",
  "formats": ["markdown", "screenshot"],
  "use_browser": "auto",
  "wait_for_ms": 1000,
  "timeout_seconds": 30
}

Bee Engine keeps one Chromium process per service and creates a fresh browser context for each request. BEE_ENGINE_MAX_PAGES controls the maximum number of rendered pages handled concurrently; the default is 4. Screenshots require browser rendering. The service blocks common media and analytics resources by default to keep page loads predictable.