Privacy policy

ScrollShot is built so that your screenshots never have to leave your computer. This page explains exactly what the extension and the web app do with data.

Last updated: October 2, 2026

The short version

No account. No sign-up. No cookies for tracking.
Screenshots are captured, edited, converted and OCR'd on your device. We do not run servers that receive your images or text.
We do not collect analytics, telemetry or crash reports.

What the extension can access

The extension asks Chrome for the smallest set of permissions that still lets it work:

activeTab and scripting: read and scroll the page you are capturing, only after you click the icon, press a shortcut or use the context menu. It has no access to other tabs or to pages you have not asked it to capture.
downloads: save the image you export to your Downloads folder.
storage: keep your settings (format, filename pattern, shortcuts behaviour). Chrome may sync these with your Google account if you have extension sync enabled; they contain no page content.
contextMenus: add the ScrollShot entries to the right-click menu.
Access to the ScrollShot web app origin so that "Open in ScrollShot" can hand the image to the workspace tab you opened. The image is passed inside your browser, not over the network.

What the web app stores

History: images you open in the workspace are kept in your browser's IndexedDB on this device (the most recent 60). You can delete any entry, or everything, from the History page. Clearing site data in your browser also removes them.
Preferences: theme and panel layout, in localStorage.
Offline files: the app shell, fonts and the OCR engine are cached by a service worker so the app works offline.

Third parties

Fonts are served from this site. The only outside resources the web app loads are the tesseract.js OCR engine and its language files (from jsDelivr and tessdata.projectnaptha.com), fetched the first time you use on-device text extraction. Those requests reveal your IP address to the CDN as any web download does; they never include your images or text. Everything is cached locally afterwards.

Optional cloud text recognition

Text extraction runs on your device by default. Where we offer it, you can choose a cloud engine (OCR.space, Google Cloud Vision or Azure AI Vision) for better accuracy. Only then, and only for that one action, is the image you are extracting text from sent to that provider, either directly from your browser or through a small ScrollShot relay that adds our access key and forwards the request without storing anything. Providers process the image under their own terms; we do not keep images, text or any record of what was recognised.

Children

ScrollShot does not knowingly collect any personal information, from anyone, of any age.

Changes

If this policy changes, the date above changes with it and the summary stays honest. Should we ever add a feature that needs an account or sends your captures anywhere, it will be opt-in and documented here before it ships.

Contact

Questions about privacy: email us or open an issue on GitHub.