Privacy Policy

Last updated: 2026-06-07

Stashed is local-first. By default, everything you save lives in your browser's local storage and never leaves your device. Some optional, clearly-labeled features (cloud sync, page archive, and AI tagging) do send data off your device — this policy explains exactly what, when, and to whom.

We have no analytics, telemetry, crash reporters, or advertising trackers. We never sell your data.

1. If you don't sign in (default)

Stashed works fully without an account. In this mode:

That's the entire footprint for signed-out use.

2. If you sign in (Google)

Signing in is optional and is used for the trial, subscription, and cloud features. We authenticate with Google via Chrome's identity API and receive a standard OpenID token. From it we store, on our backend:

Signing in alone does not upload your saved pages. Your stash stays local unless you have an active Pro subscription or trial (see below).

Our backend is Convex (convex.dev), which hosts our database and server functions. Data is scoped to your account.

3. Pro / trial features that send data off your device

These features are only active for Pro or trial users, and only for pages you choose to save.

Cloud sync

When you sync, your stash (page URLs, titles, tags, notes, collections, and save timestamps) is uploaded to our Convex backend so it can be merged across your devices. Deletions are soft ("Trash") and are removed permanently after 30 days. This data is stored per-account on our servers; it is not end-to-end encrypted, because the archive and search features below require us to process it server-side.

Page archive

For pages you save, Stashed captures the readable text of the page (capped in size) and stores it on our backend so the saved copy survives if the original link dies, and so you can search inside saved pages. Only the extracted text is stored — we do not capture passwords, form inputs, or content behind logins beyond what is visible as page text at save time.

AI tagging

When you save a page, its title, URL, and a truncated excerpt of its text are sent — through our Convex backend — to a third-party AI provider (currently NVIDIA's hosted models) to generate topic tags. The provider processes this content to return tags; their handling of that data is governed by their own terms. We store only the resulting tags, not a separate copy of the request. This call happens only for Pro/trial users and is rate-limited per account.

4. Payments

Subscriptions are handled by Creem (creem.io), our Merchant of Record. Creem processes your payment details directly — we never see or store your card information. We retain only Creem's customer and subscription identifiers and your subscription status. See Creem's privacy policy for how they handle payment data.

Cancelling stops renewal; Pro remains active until the end of the paid period.

5. Your data, your control

6. Children

Stashed is not directed to children under 13 and we do not knowingly collect their data.

7. Changes

We will update this policy and its "Last updated" date before any new data-handling behavior ships.

8. Contact

Questions, concerns, or data deletion requests: