Privacy Policy
Last updated: June 14, 2026
The OCR Buddy extension runs entirely on your device: no servers, no accounts, no telemetry. It does not collect, transmit, sell, or share any of your data. The website (ocr-buddy.com) uses one privacy-friendly, cookieless analytics service for aggregate visit counts — and you can opt out. Details in §4.
1. Who this covers
This policy applies to the OCR Buddy Chrome extension and the website ocr-buddy.com. OCR Buddy is free, open-source software released under the MIT license; its full source is on GitHub.
2. What the extension processes
When you select a region of a page (or open, paste, drop, or right-click an image), OCR Buddy captures that image and recognizes the text in it. This happens entirely on your device, using OCR models running locally inside the extension. Nothing is ever logged or transmitted anywhere. If the optional capture history is enabled (it can be turned off or cleared in the panel at any time), the last 20 captures and their results are kept in the extension's local storage on your device — and nowhere else.
| Data | Where it goes | Stored? |
|---|---|---|
| Captured screen region (image) | Processed locally, in memory | Only if history is on — locally only |
| Recognized text / LaTeX / Markdown | Shown to you in the side panel; copied only when you click Copy | Only if history is on — locally only |
| Capture history (last 20 captures + results) | Saved locally via chrome.storage.local; you can turn it off or clear it in the panel at any time | Locally only |
| Your settings (capture mode, language, Prose/Code view) | Saved locally via chrome.storage.local on your device | Locally only |
3. What we collect
The extension: nothing. The OCR Buddy browser extension has no analytics, no tracking, no cookies, no account, and no remote server. We never see what you capture, and we have no way to.
The website: cookieless analytics only. ocr-buddy.com uses a privacy-friendly, cookieless analytics service for aggregate visit counts — no cookies and no personal data, and you can opt out. Details in §4, Website analytics.
4. Website analytics
This website (ocr-buddy.com) uses Vercel Web Analytics, a privacy-friendly, cookieless service, to count page views and see which pages and links people find useful. It is the only third-party service the site uses, and it applies to the website only — the OCR Buddy extension contains no analytics and never connects to it.
- No cookies and no tracking storage. Nothing is written to your device to identify you across visits.
- Aggregate, non-identifying data only: page views, the referring site, and coarse details such as country, device type, operating system and browser. It is not used to build a profile of you.
- No stored identifiers. Visitors are counted with a hash that Vercel derives from request data and rotates daily; your IP address is not stored.
- You can opt out. Use the “Opt out” button in the notice at the bottom of the page. Your choice is kept locally and stops analytics from loading on the pages you visit afterwards.
Vercel processes this data as our service provider; see Vercel's privacy policy and its analytics privacy notice. Because the analytics are cookieless and collect no personal data, no consent banner is legally required; we show the notice and offer opt-out anyway, to stay upfront.
5. Network use
OCR Buddy performs OCR with no network connection at all — the bundled models ship inside the extension, so the default experience works fully offline. There are exactly two network activities, neither of which carries any of your data: Chrome downloading the extension (and its updates) from the Chrome Web Store, governed by Google's privacy policy; and, only if you explicitly select a non-Latin language pack, a one-time download of that model file from the pinned open-source model repository on GitHub. The request contains nothing but the file URL — no page content, no images, no identifiers from the extension — and the file is cached locally so it is never fetched again.
6. Permissions and why they are needed
- activeTab — to screenshot the visible tab where you start a capture, only after you act.
- scripting — to draw the selection overlay on the page when you start a capture.
- offscreen — to run the OCR model in a background document, off the service worker.
- sidePanel — to show the result in Chrome's side panel.
- storage / unlimitedStorage — to remember your settings, your local capture history (if enabled), and to cache OCR model files on your device.
- contextMenus — to offer "OCR this image" when you right-click an image. The image is processed locally like any capture.
- Optional site access — requested per-site, at the moment you try to capture a tab you didn't invoke OCR Buddy on. The screenshot is gated behind your explicit consent.
None of these permissions are used to collect or send data off your device.
7. Data sharing
We do not sell, rent, trade, or share your personal data with anyone. The only third party involved is Vercel, which hosts the website and provides the cookieless analytics described in §4, acting as our service provider. The extension shares nothing, because it collects nothing.
8. Children
OCR Buddy is a general-purpose utility and is not directed at children under 13. It collects no personal information from anyone.
9. Changes to this policy
If this policy changes, the updated version will be posted at this URL with a new "Last updated" date. Material changes will be noted in the extension's release notes on GitHub.
10. Contact
Questions about privacy? Open an issue on GitHub or email contact@ocr-buddy.com.