agent-browser-cli

You can automate web testing and form interaction by using supervised Chromium browser controls to navigate pages, inspect actionable elements, fill forms, and click controls without requiring Selenium or Playwright. It safely preserves browser and profile identity during operations.

Yes, you can take screenshots and save PDFs during browser automation. The skill supports diagnostics and evidence-capture tasks, allowing you to scan page content, capture visual evidence, and safely handle the resulting binary artifacts.

To inspect network requests and console logs for web troubleshooting, you can use the skill's diagnostics features to monitor network activity, review console logs, and inspect cookies alongside the CDP state within a supervised Chromium environment.

No, this browser automation approach does not require Selenium or Playwright. The skill enables reliable browser perception and interaction independently, utilizing high-level CLI workflows to control supervised Chromium browsers directly.

The best way to handle stale elements during web page navigation is to refresh element snapshots after page changes. The skill applies high-level CLI workflows that require explicit tab and profile identity to ensure accurate element targeting.

When managing multiple tabs during browser automation, you must maintain explicit tab and profile identity. The skill requires safe handling of sensitive browser data and explicit snapshot refreshing after any page changes to avoid operational errors.