Privacy policy

Effective 9 October 2026

In short

• There is no sign-up: we never ask for your name, email or phone number.
• Your books, bookmarks and words stay on your phone.
• Only what you ask to translate reaches our server: a word with its sentence, a paragraph or a page.
• AI models receive that text without your device number, IP address or anything else about you.
• No ads, no trackers, no selling of data.

1. 1. Who we are
2. 2. What stays on your device
3. 3. What our server receives
4. 4. Who we share data with
5. 5. The sl-reader.app website
6. 6. How long we keep it
7. 7. Legal basis and location
8. 8. Your rights
9. 9. Children
10. 10. Security
11. 11. Changes
12. 12. Contact

1. Who we are

SLReader is an app for reading in a foreign language, developed by Dmitrii Golovin (“we”). This policy explains what data the SLReader app and the sl-reader.app website receive, why they need it and how long it is kept. Questions go to support@sl-reader.app.

2. What stays on your device

The books you open in SLReader, your reading progress, bookmarks, the Words list, settings, uploaded fonts and swapped pages are stored only on your phone. We do not see them and never receive them.

If you add your own OpenAI key, it is kept in the system’s secure storage (the Keychain on iPhone, the Keystore on Android). Delete the app and all of this goes with it.

3. What our server receives

A device number and technical details. On first launch the app creates a random device number. It is not linked to your name, email, or your Apple or Google account. Along with it the server receives the platform (iOS or Android), the phone model, the system and app versions, the system language and region, the time zone and the interface language. We use the number to count your daily limit and to find your device when you contact support.

IP address. The server sees it with every request. We keep the device’s most recent IP address and the country we derive from it with the DB-IP database, on the server itself, without calling any outside service. In the request log the IP address is stored without its last part (for example, 203.0.113.x).

Text to translate. When you tap a word, translate a paragraph or a selection, or turn on word swap, the app sends the text needed to the server: a word together with its sentence, a paragraph or a page. Occasionally, to detect a book’s language, it sends a few short excerpts from it. This may be text from your own book. The server passes the text to an AI model (section 4) and keeps the answer in a shared cache, so the same request is served instantly next time. Sentences and paragraphs themselves are not stored in the cache — only their fingerprint (a hash) and the finished translation. For words and expressions, the word itself is stored.

Request log. For every translation request we record the device number, the time, the kind of request, which model answered, what it cost, whether the answer came from the cache, and which answer was returned. We need this for the daily limit, for keeping track of costs and for finding bugs. Separately we keep daily totals: how many requests of each kind a device made that day, and what they cost.

Server logs. The server records every request, from the app and to the website: the time, the address requested, the full IP address and, for the app’s requests, the device number, along with any error that occurred. Of what you send for translation, only the word or expression itself can appear there: sentences, paragraphs, pages and support messages are not recorded. We need these logs to investigate failures and to protect the service from abuse.

Support messages. If you write to us from the app, we receive the message, the device number, the app version, the phone model, the system version and the interface language. Our reply is stored alongside it and shows up in the app.

Your own OpenAI key. If you add your own key, the app sends it with each request. The server uses it to call OpenAI on your behalf and does not store it. Those requests are billed to your OpenAI account, and OpenAI’s policy applies to them.

Stories and classics. The app downloads catalogue texts and covers from our server. We do not keep track of what you read.

4. Who we share data with

• AI model providers: OpenAI, Anthropic, Google and OpenRouter (USA). They receive only the text to translate and technical parameters such as the languages — without your device number, IP address or anything else about you — and process it under the terms of their APIs.
• Hosting: Hetzner Online GmbH (Germany). Our server runs and our database is stored on its machines.
• Crash reports: Sentry (Functional Software, Inc., USA), which stores our reports in the EU (Germany). When the app crashes or the server hits an error, a report goes there: what failed and where in the code, the platform, the phone model, the system and app versions. It does not receive your device number or IP address, and we set it up so that the text you read does not reach it.
• Apple and Google distribute the app. If you allowed sharing analytics with developers in your phone’s settings, they pass us anonymised statistics and crash reports.
• When the law requires it, or when a government authority makes a binding request.

We do not sell data, show ads, or use third-party analytics or trackers.

5. The sl-reader.app website

The website sets no cookies, uses no analytics or counters, and loads no third-party fonts or scripts. The server keeps ordinary access logs (IP address, time, page address) for up to 40 days, for security and diagnostics.

6. How long we keep it

• The request log: up to 13 months.
• Server logs: up to 40 days.
• Crash reports at Sentry: up to 90 days.
• The device number, technical details and daily totals: for as long as you use the app, or until you ask us to delete them.
• Support messages: for as long as they are needed to answer you and keep the history of your requests. We delete them on request.
• The translation cache: indefinitely. Once the request log is cleared, it is not linked to any device.
• Database backups: for a limited time; older copies are deleted as new ones are made.

7. Legal basis and location

We process data to do what you ask (translate text, answer your message) and in our legitimate interests: to enforce daily limits, protect the service from abuse and fix bugs. The server and crash reports are located in Germany. Text to translate is processed by AI model providers in the USA.

8. Your rights

You can find out what data we hold about your device, have it corrected or deleted, and object to its processing. Write to support@sl-reader.app or use “Report a problem” in the app, and include your device number — you’ll find it under Settings → About. There is no sign-up, so this number is how we find your data.

You also have the right to lodge a complaint with the data protection authority of your country — in the European Union, your national supervisory authority.

9. Children

SLReader is not directed at children under 13, and we do not knowingly collect their data. If you believe a child has given us any data, write to us and we will delete it.

10. Security

The app and the website talk to the server over HTTPS only. A device’s access key is stored on the server as a hash, readers’ OpenAI keys are never stored, and only the developer has access to the database.

11. Changes

If this policy changes, we will publish the new version on this page with a new date, and tell you about significant changes in the app.

12. Contact

support@sl-reader.app, or Settings → About → Report a problem in the app.