GrammarMe / Privacy
Privacy Policy
Effective date: June 16, 2026 · Last updated: June 16, 2026
In short: GrammarMe processes all of your text locally, on your own device. It does not collect, store, or transmit the content you translate or proofread, and it contains no analytics, telemetry, or advertising.
1. Who this applies to
This policy describes how the GrammarMe browser extension (“GrammarMe”, “the extension”, “we”) handles information. It applies to the extension itself. It does not apply to third-party websites you visit, or to services whose terms are described below (Google Chrome's built-in AI and the Hugging Face Hub).
2. Data we do not collect
GrammarMe does not collect, store, or transmit any of the following:
- The text you translate or proofread, or its results.
- Your browsing history or the contents of pages you visit.
- Personal information such as your name, email address, or location.
- Analytics, usage metrics, telemetry, or crash reports.
- Advertising or cross-site tracking identifiers.
GrammarMe has no account system and no server backend that receives your data.
3. On-device processing
All translation, language detection, and grammar/spelling correction run locally on your deviceusing Chrome's built-in AI (Gemini Nano), through the browser's Translator, Proofreader, and Prompt APIs. The text you select and the results produced from it stay on your device and are never sent to us or to any third party.
4. Settings stored on your device
GrammarMe saves only your preferences — such as your target language, the languages you write in, and which AI engines/models to use. These are stored with Chrome's chrome.storage.sync API.
Because chrome.storage.sync is used, Chrome may synchronize these preference values across browsers where you are signed in to your Google Account. That synchronization is performed by Google and governed by Google's Privacy Policy, not by GrammarMe. Only your preference settings are involved — none of the text you translate or proofread is ever stored or synced.
5. Optional offline model downloads
If Chrome's built-in AI is unavailable, you can optionally enable offline fallback engines. When you do, the extension downloads machine-learning model files from the Hugging Face Hub (https://huggingface.co/*).
This is a download of files to your device. The text you translate or proofread is not sent to Hugging Face — inference still runs locally. As with any file download, Hugging Face may log standard request metadata such as your IP address, subject to Hugging Face's privacy policy. No model download occurs unless you choose to enable an offline engine.
6. Permissions and why they are used
storage— to save your preference settings (and optionally sync them via Chrome, as described above).offscreen— to run on-device AI/model inference in a background document.activeTab— to read the text you have selected on the current tab, only when you invoke the extension.host_permissionsforhttps://huggingface.co/*— to download optional offline model files, only when you enable an offline engine.
7. No sale or sharing of data
We do not sell, rent, trade, or share your data with anyone. We do not track you across websites and we do not build advertising or behavioral profiles. GrammarMe is not monetized through your data.
8. Data retention
Your preference settings remain in your browser until you change them or uninstall the extension. Uninstalling GrammarMe removes its locally stored data. Because there is no server-side storage, there is no remote data for us to retain or delete.
9. Children's privacy
GrammarMe is not directed to children under 13 and does not knowingly collect personal information from anyone.
10. Changes to this policy
We may update this policy from time to time. Any changes will be posted on this page with an updated effective date. Material changes will be reflected here before they take effect.
11. Contact
Questions about this policy or GrammarMe's data practices? Reach out at joseluiskoller98@gmail.com or open an issue on the GitHub repository.