Privacy Policy

Note for review: the [[PENDING]] items must be completed by the owner. The processors and international-transfer sections must be checked against each provider’s current contract before publishing.

1. Data controller

David Latorre Gomis, tax ID 33488436E, registered at Carrer Torres Quevedo, 112, 1º, 3, 03204 Elche, Spain. Privacy contact: privacidad@tagsora.com.

2. What data we process

Account data. Email address, hashed password, language and app preferences, and a public username if you choose one.

Content you save. The URLs you save and their metadata (title, description, author, image), the readable text extracted from those pages, your tags, notes, highlights and collections, and the vector representations (embeddings) that make meaning-based search possible.

Service usage data. Counters of AI feature consumption (summaries, searches, tagging, assistant) needed to apply your plan’s limits, and technical error logs.

Billing data. If you subscribe to premium, payment is processed by [[PAYMENT PROVIDER]]. We never receive or store your card details. We only receive the subscription identifier and its status.

Inbound email. If you enable the save-by-email address, we process the messages you forward to it in order to extract their links.

Sources and listening. Which RSS sources, podcasts and stations you follow, which of their items you have seen or saved, and your listening or reading position in each. The sources themselves (the feed address and its articles or episodes) are shared, non-personal data.

Profile and sharing. Your public username, whether you appear in the people search, the collections you share or publish, your public highlights and your comments and reactions in shared collections. If you invite someone to a collection by email, we keep that address until they accept or the invitation is withdrawn.

Uploaded files (Premium). The documents (PDF, ePub, Word, LibreOffice, text, Markdown, saved web pages) and images (photos and screenshots) you upload, with the name you give them, their size and the date. They are private: only you see them unless you choose to share them in a Space or an invite-only collection; they are never published.

• Images: before saving them we remove their hidden data (EXIF, XMP, IPTC), including the GPS location, the date and the phone model; we only keep the orientation, so they display upright. So that you can find them when you search, each image is analyzed once with Google’s paid Gemini API: it produces a description, the text that appears in it and some tags. We ask the model not to identify people by their face or infer personal details that are not written in the image. [[VERIFY WITH LEGAL REVIEW: photos of people or special categories of data]]
• Shared processing: if someone else uploads exactly the same file (the same bytes), a single copy is kept and its automatic analysis is reused. Your highlights, notes, tags and the name you give it are always yours.

Imports. Export files from other services that you upload are processed to extract the URLs and their tags; the file itself is not kept.

Premium availability notice. If you tap “Notify me when it can be purchased”, we store the plan you chose, the screen you asked from and the date, solely to email you once purchasing opens.

Terms acceptance record. Which version of the terms you accepted, when and from which platform.

Blocklist of banned accounts. If an account is banned for a serious breach of the community guidelines, we keep an encrypted fingerprint (a SHA-256 hash) of its email address, not the address itself, along with a reason code and the date. Its only use is to reject a new sign-up with that same email.

Content reports. If you report a public page (a collection, a profile or a compilation), we store which page it is, the type of report, the reason you write and the date. If you’re signed in, the report is linked to your account; if not, you can leave an optional email so we can tell you what we decided. To prevent abuse of the form we use an encrypted fingerprint of your IP address (not the address itself), deleted after two days. If the reported content is yours, the report is linked to your account and we’ll write to you if we take any action.

Diagnostics. App error reports (error type, version, device model). Before they are sent, the addresses of the pages you save, search terms and any request parameters are stripped; they never contain your library.

3. Purposes and legal bases

We do not advertise, we do not profile for commercial purposes and we do not sell data to anyone. “For you” recommendations are computed only from your own library and the people you follow, to order content inside the app; they are not used for anything else.

4. Who it is shared with (processors)

Each one processes data on our behalf, under a data processing agreement, and only for the purposes described.

Services that receive queries, but no account data. To search podcasts and stations, our server queries the public Podcast Index and Radio Browser indexes with the term you type; they receive neither who you are nor what you save. RSS feeds are fetched from each publisher’s server, which sees Tagsora’s request, not yours.

Third-party content loaded on your device. If you play an embedded YouTube video or Spotify item, or listen to a podcast or a station, your device connects directly to that third party’s server and its own policies apply. YouTube videos load in privacy-enhanced mode (youtube-nocookie.com).

5. International transfers

Some providers are located outside the European Economic Area. In those cases the transfer relies on [[STANDARD CONTRACTUAL CLAUSES / ADEQUACY DECISION — VERIFY WITH EACH PROVIDER]].

6. How long we keep data

• Your content: for as long as your account is active.
• On account deletion: your library, notes and highlights are removed. Deletion takes effect immediately in the database; backups are overwritten within [[PERIOD]] days at most.
• Billing data: for as long as tax law requires ([[PERIOD — typically 4 years in Spain]]).
• Articles and episodes of the sources you follow: shared data, kept for 180 days or the most recent 200 per source; what you save to your library does not expire.
• Uploaded files: for as long as they are in your library. If you stop being Premium they are kept and you can still read them. They are deleted when you remove them, when you delete your account or, if Premium has expired and the account has been inactive for 12 months, after emailing you 30 and 7 days before. The original and its analysis are deleted once nobody keeps that same file.
• Export files: 24 hours after they are generated; then they are deleted from storage.
• Premium availability notice: until the email is sent, until you withdraw it or until you delete the account, whichever comes first.
• Terms acceptance record: for as long as the account exists and, after deletion, the limitation period set by law.
• Blocklist of banned accounts: 5 years from the ban, unless a different period is set when it is imposed. [[PERIOD — to review with the lawyer]]
• Content reports: one year after they’re closed. The reporter’s email and account are deleted sooner: six months after closing. The IP address fingerprint, after two days.
• Error reports: [[PERIOD — the retention configured in Sentry, 90 days by default]].

Note that the text extracted from a web page is shared and non-personal: it is stored once per address, not once per user. Deleting your account removes your bookmark, your notes and your highlights; it does not remove that shared text, which cannot identify you.

7. Your rights

You can exercise your rights of access, rectification, erasure, objection, restriction and portability, and withdraw consent where processing is based on it.

Two of them need no email at all: inside the app you have a full export of your data in a reusable format and account deletion, both under Settings.

For the rest, write to privacidad@tagsora.com. If you believe we have not handled your request properly, you can complain to the Spanish Data Protection Agency (www.aepd.es) or to your local supervisory authority.

8. Cookies and analytics

This site uses no tracking or advertising cookies, which is why you are not seeing a consent banner. Traffic measurement uses Cloudflare Web Analytics, which sets no cookies and builds no visitor profile.

The app does store on your device the technical information needed to keep you signed in and remember your preferences. That is not tracking: without it you could not stay logged in. As of this version the app includes no product analytics; if it is added, it will not collect the content of what you read or save, and it will be reflected here.

9. Security

Data access is restricted per user in the database itself (row level security): every row belongs to an account and the server does not return other accounts’ rows. Connections are encrypted, and the AI providers’ keys live only on the server, never on your device.

10. Children

The service is not directed at children under 16 and we do not knowingly collect data from them.

11. Changes to this policy

If it changes substantially, we will give notice by email or inside the app before it takes effect, and update the date above.