Last updated: 2026-06-26 · Dolphin Studio LLC, Wisconsin, USA
Short version: Snap OCR runs 100% on your own device. It collects nothing, sends nothing, makes no network requests, and the pixels it reads never leave your browser.
When you click the Snap OCR toolbar icon and drag a box over part of a page, the extension reads that region and recognizes the text inside it. All of the optical character recognition (OCR) happens locally in your browser using a text-recognition engine (Tesseract / WebAssembly) that is bundled inside the extension. The recognized text is shown to you and copied to your clipboard. That is all it does.
None. Snap OCR has no servers, no account, no sign-in, no sync, and no backend. We do not collect, transmit, store, sell, or share any of your data — not the images you capture, not the recognized text, and no personal or usage data. We cannot see any of it.
No. The captured pixels and the recognized text are processed entirely on your device and never leave the browser. Snap OCR makes no network requests of any kind — it works fully offline, including the OCR engine and its language data, which ship inside the extension. There are no external services, CDNs, or API calls.
Snap OCR uses activeTab and scripting so that, only after you click its
icon, it can draw the selection overlay on the current tab and capture the region you drag over.
clipboardWrite is used to copy the recognized text to your clipboard. offscreen
hosts the local OCR engine. None of these involve any network access or data collection.
There is nothing stored to remove — Snap OCR keeps no history. Uninstalling the extension removes it completely.
None. No analytics, no ads, no trackers, no external SDKs, no remote code.
Questions: dolphinstudio101@gmail.com