Privacy policy

Last updated: September 30, 2026

In two lines

Biblo runs entirely on your computer. We run no servers: what you do in the app is not collected, not sent to us, not stored by us and not visible to us. No tracking, no usage statistics, no account.

The same goes for this website: no profiling cookies, no third-party analytics. To show the current version and the download count, it queries GitHub's public API straight from your browser.

1. Who provides the app

Biblo is provided by Lorenzo Montoli, a private individual, free of charge and with no commercial activity behind it. You can write to him at lorenzomontoli@gmail.com.

2. What stays on your computer

Everything the app creates or manages lives on your computer, under your control — and deleting it is entirely up to you, from inside the app or by removing its data folder:

• your documents and the search index built from them;
• your chat history;
• your settings;
• your API keys, saved in the app's local configuration.

3. What leaves your computer, and when

When you use the AI features (chat, document inspection, connections), the app sends your question and the excerpts of your documents needed to answer it to the model provider, Google Gemini (generativelanguage.googleapis.com). The requests use your own API key, under your own account, and travel straight from your computer to the provider: we never see them, receive them or store them. That relationship is between you and the provider, governed by its own terms and privacy policy — choose what you send accordingly.

Be aware of how the provider's free tier treats your data:

• Google Gemini (free tier): human reviewers may read what you send and what the model answers, and the use of your data to train models cannot be turned off. Google asks you not to submit confidential or personal information on this tier.

While inspecting your documents, the app also queries Wikipedia (<language>.wikipedia.org) for the short description of the people, works and concepts cited in them, which it uses to build the cards. What travels is the term found in your document — never the document, the excerpts, or your questions. These requests are anonymous (no account, no identifier, no cookie), they go straight from your computer to Wikimedia, and their results are cached on your computer so the same term is not looked up twice.

4. Other connections

The app also contacts GitHub, and these connections carry no content from your documents: to check its public API for a newer version; to download an update, when you accept one, from the public repository LorenzoMontoli/biblo-releases; and to download the embedding model from there if it is not already on your computer — in the current versions the model ships inside the installer, so this download normally does not happen. These requests carry no identifier beyond your IP address, which reaches GitHub — not us.

5. The GDPR in short

• If you use Biblo for purely personal or household purposes, the GDPR does not apply to that use at all (Art. 2(2)(c)).
• If you use it professionally on documents containing other people’s data, you are the controller of that processing, including whatever you send to the model provider.
• We are neither controller nor processor for the content of your documents, questions and answers, for a simple reason: we never receive any of it.
• The only personal data we process is the email you choose to send us for support, used solely to reply to you.

6. Your rights, changes, contact

Your data stays on your device, so you can consult, correct or delete it yourself at any time. For support correspondence, or for any privacy question, write to lorenzomontoli@gmail.com; you can also lodge a complaint with a supervisory authority (in Italy, the Garante per la protezione dei dati personali).

If this policy changes in a meaningful way, we increase its version number and the app asks you to read and accept it again.