Chrome Extension¶
The VisualRunner extension has two separate modes — pick the one that fits what you need.
Free Local mode — no account needed¶
Capture screenshots directly from the extension's side panel with no sign-in and nothing uploaded anywhere. Everything — your captures, your custom country profiles, your language history — stays in your browser and exports as a zip file you download yourself.
Free Local mode has its own limits, separate from any paid plan:
| Limit | Value |
|---|---|
| Languages per session | 5 |
| Country profiles | 2 |
| Viewports | 2 |
| Screenshots per session | 20 |
Good for a quick one-off check, or trying VisualRunner out before creating an account.
Cloud mode — signed in, connected to your workspace¶
Sign in from the extension's side panel (this is a separate sign-in from the dashboard — signing in on the website doesn't automatically sign in the extension, and vice versa). Once signed in, the extension is what actually powers manual captures started from the Captures & Comparisons tab in your dashboard — the dashboard hands off the work to your installed extension, which does the capturing and uploads results to your project.
The first time you run a cloud capture, you'll be asked to confirm a one-time consent to uploading screenshots to VisualRunner's cloud — this is separate from Free Local mode's captures, which are never uploaded.
Installing¶
Install from the Chrome Web Store. VisualRunner requires a Chromium-based browser (Chrome, Edge, Brave, etc.) — it won't work in Firefox or Safari.
Permissions¶
The extension asks for broad page-access permission only when you actually start a capture, scoped to just the sites that specific capture run needs — not for every website up front at install time. It also needs standard extension permissions to control tabs, take screenshots, store data locally, and show notifications.
Detecting languages and pages¶
When you point the extension at a page, it looks at several signals to suggest which language
you're looking at — hreflang tags, a visible language switcher, the page's declared language, the
URL itself, and anything you've confirmed before on that site — rather than guessing from just one
of them.