Privacy Policy for Research Sync

Last updated: July 9, 2026

Research Sync (“the app”) is designed to respect your privacy. This document explains what data the app handles, where it is stored, and how it is used.

Summary

Research Sync does not collect personal information about you. The data it stores is the URLs and page titles you explicitly choose to save, along with read/unread status and timestamps, plus — for articles you save for offline reading — the readable text of those articles. That data is stored on your own devices and synced between them via a small cloud relay operated by the developer, solely so your saved list stays up to date across your iPhone, iPad, and browser. Offline article text is end-to-end encrypted on your device before it is synced, so the cloud relay only ever holds unreadable ciphertext, and it is never shared with anyone. To sort your saved links into folders automatically, the title and URL of a saved link — never the article text, and no information about you — are sent to Google’s Gemini API to suggest a folder name. That is the only third party involved, it is used solely to categorize your links, and under Google’s paid API terms this data is not used to train Google’s models. Your data is never sold.

Data the app stores

When you save a page, the app stores:

• The URL of the page
• The title of the page (typically the <title> of the HTML document)
• A timestamp for when you saved it
• A timestamp for when you last changed it
• A read/unread flag
• A flag marking whether you've deleted it, so the deletion can sync to your other devices
• Any optional notes / context that you add to the notes field for a URL
• A short folder name used to group the item, suggested automatically (see “AI folder organization” below)
• For any article you save for offline reading: the readable text of that article (extracted from the page and cleaned up for reading), so it stays available without a network connection. This text is kept in a local on-device cache, and — if synced — is end-to-end encrypted on your device before upload, so the cloud relay only stores ciphertext.

That is the entirety of what is stored. The app does not collect or transmit:

• Your name, email address, or any account information
• Device identifiers, advertising identifiers, or IP addresses
• Your browsing history outside the pages you choose to save
• Location data
• Analytics, telemetry, or crash reports of any kind
• Contacts, calendar, photos, or any other on-device data

Where the data is stored

Saved items are stored in two places:

1. A local container on each device (an App Group container on iOS, or the browser’s local extension storage) so the app and any browser extension can read and write to the list immediately, even offline. The readable text of articles you save for offline reading is kept in a separate on-device cache on the same device.
2. A cloud sync store, hosted on Cloudflare Workers and operated by the developer, used to relay your saved list — and the encrypted text of your offline articles — between your own devices, and to request folder suggestions as described under “AI folder organization” below. Access requires a private sync key that is generated on your device; the same key is also used to end-to-end encrypt the text of offline articles before it is uploaded, so the store holds only ciphertext for that content. Aside from the automated folder categorization, the developer does not separately access, read, or analyze your data.

The only network calls the app and extension make are to this cloud sync service, to push and pull your own saved list. The sync service in turn forwards the title and URL of saved links to Google’s Gemini API for folder categorization, as described under “AI folder organization” below; the app and extension never contact Google directly, and the article text of offline saves is never sent.

AI folder organization

To keep your saved list tidy, the app can automatically sort saved links into folders (such as “Tech,” “Recipes,” or “News”). This works as follows:

• The cloud sync service sends the title and URL of links that do not yet have a folder to Google’s Gemini API, which suggests a short folder name for each. The suggested name syncs back to your devices like any other field.
• Only the title and URL are sent. The readable text of your offline articles is never sent to Google — it stays end-to-end encrypted and is never transmitted in a form anyone but you can read. No account information, device identifiers, or information about you is sent, because the app does not hold any.
• Google acts only as a processor for this categorization. Under the terms of Google’s paid Gemini API, the data sent is not used to train Google’s models and is not used for advertising.
• Folders are only ever suggested for links that do not already have one; the feature never re-reads or re-categorizes links you have already filed, and a folder you set yourself is never overwritten.

Browser extension

An optional Chromium browser extension (for Dia, Chrome, Edge, Arc, Brave, etc.) is available separately. It stores the same fields (URL, title, timestamps, read flag, deleted flag, notes) using the browser’s local extension storage, and syncs them with the same cloud sync service described above. When you save an article for offline reading, the extension reads the content of that page — only at the moment you choose to save it — using bundled, on-device reader libraries, to extract the article text for the offline copy. It does not read or monitor pages you have not chosen to save, and it does no background browsing.

Permissions the iOS app requests

• Network access — to sync your saved list with the cloud sync service described above, across your own devices.
• Share Extension activation — when you choose Research Sync from the iOS share sheet, the extension reads the URL being shared so it can save it. It does not read anything else.
• Safari Web Extension permissions — when enabled in Safari, the extension reads the URL and title of the page you are currently looking at, but only when you tap its toolbar icon.
• Siri / Shortcuts intent — when you invoke the app’s save action via Siri or Shortcuts, the URL you provide is saved.

Third parties

Research Sync does not use any third-party analytics, advertising, crash-reporting, attribution, or tracking SDKs.

The one third party involved is Google, whose Gemini API is used solely to suggest folder names from the title and URL of your saved links, as described under “AI folder organization” above. This is the only case in which any of your data is shared with a third party, it is limited to titles and URLs (never article text or any information about you), and under Google’s paid API terms it is not used to train Google’s models or for advertising. No other third party has access to your data.

Your control over your data

You control your saved list completely:

• Delete an individual saved item from within the app or extension at any time — the deletion syncs to your other devices.
• Uninstall the app, which removes all locally stored data on that device.
• To request removal of your data from the cloud sync store, contact the developer at the email address below.

Children’s privacy

The app is not directed to children under 13 and does not knowingly collect any personal information from anyone, regardless of age.

Changes to this policy

If this policy changes in a meaningful way, the updated version will be posted at this URL and the “Last updated” date at the top will be revised.

Contact

Questions or concerns about this privacy policy: shearm@mac.com