Browserless hosts managed headless browsers so you don't have to run Chrome yourself. Point existing Puppeteer, Playwright or Selenium code at a Browserless WebSocket endpoint and it runs there.
It also offers BrowserQL, a GraphQL language for automation with stealth and structured extraction, REST endpoints for screenshots, PDFs and scraping, and an MCP server with scrape and crawl tools for agents.
There is a free account, and the open-source core is on GitHub so you can self-host. Docs are at docs.browserless.io.