Your reading, explained.

Browsing and original blogs

You can browse collected publisher metadata without creating a reading profile. Readwell stores titles, publisher excerpts, dates, source links, and inferred topics. Full articles stay with their publishers. Their sites apply their own privacy policies when you visit them.

Accounts and personal data

Clerk provides sign-in and session management. Readwell stores your reading profile under your verified account identifier in its application database and keeps an account-specific device copy for interrupted saves. Your most recently opened publisher blog is stored privately so you can return to it; opening a blog does not mark it as read. Your name, bio, follows, interests, custom feeds, bookmarks, read logs, activity days, and recommendation feedback are personal account data.

A public bookmark or read-log contribution stores a hashed reader identifier, article identifier, action type, active state, and server timestamp. Hashing permits deduplication; it does not make the record anonymous to this application. Public pages show aggregate totals.

Recommendations and retention

Recommendations use publisher metadata and your reading choices. Embeddings are computed from public titles and excerpts; private account profiles are not sent to an embedding API. Private blog-issue reports store your account’s hashed identifier, the article, issue type, optional details, and submission date until you delete your reading data. Activity-day history is limited to 400 days. A hashed network or account identifier limits repeated expensive searches. Raw network addresses are not stored by that limiter; inactive limit records become eligible for cleanup after ten minutes. A hashed account reset marker prevents stale tabs from silently restoring deleted reading data. Feed editions expire after one hour and are cleaned up in bounded batches. Other reading records remain until you remove them or delete your reading data.

Account-specific browser copies remain until cleared or replaced. Deletion removes the current browser’s confirmed account cache and application database records; backups, if operated separately, require their own retention policy. This local app does not promise an external backup deletion schedule.

Your controls

Use Reading data & privacy to export or delete your app reading data, or delete both your reading data and sign-in account. Manage identity information and sign-in methods through Manage account. Clearing this browser’s site storage removes its local copies.

Automatic cleanup following deletion of the sign-in identity requires the verified Clerk deletion webhook to be connected. Reading-data deletion in Settings works directly without that integration.