Privacy Policy
Last updated: 2 August 2026
Tentaco is built privacy-first. There is no mandatory registration — you can use all tools without an account. Most tools run entirely in your browser; data stays on your device unless you explicitly use optional features such as file exchange, device sync, Image Studio AI or cloud assistant modes.
Controller
Tentaco is operated by Daniel Kuhlbarsch, Olvenstedter Grund 2, 39130 Magdeburg. Email: tentacocc@proton.me, phone: 0152 05939006.
What we collect
We do not require personal data and registration is optional. To improve Tentaco, we use privacy-friendly website analytics on our own infrastructure (see below). Contents processed in local tools are not stored on our servers unless you enable optional features described below.
Local browser storage
Favourites, recent tools, text snippets, tool collections, theme settings, assistant chat history, assistant mode preferences and workflow presets are stored locally in your browser until you delete them.
File exchange (optional)
If you use File exchange, you upload a file to our server to receive a temporary download link. We store the file name, size, MIME type and the file itself. Files are automatically deleted after at most 6 hours. We do not analyse file contents.
Assistant (local, server LLM & cloud)
By default the Tentaco assistant runs entirely in your browser — your questions are matched locally and are not sent to our servers. Chat history and mode preferences stay in your browser until you clear them. Optional server LLM: when enabled in assistant settings (and, if configured by the operator, may stay enabled until you change it), difficult questions may be sent to our infrastructure. We transmit your query and limited context: current tool or page, file name and type, a short text preview of loaded studio files (up to about 1,200 characters — not full files), recent error snippets and recent chat turns. Processing uses an open-source model on our own servers (typically EU hosting). We do not log or permanently store assistant prompts or responses. Rate limits apply (30 requests per 15 minutes per IP address). Legal basis: Art. 6(1)(a) GDPR (consent when you enable or keep server LLM enabled) and Art. 6(1)(f) GDPR (legitimate interest in rate limiting). You can disable server LLM at any time in assistant settings. Optional cloud mode (OpenAI): if you separately enable cloud mode, we send your query to OpenAI (USA) for richer answers — only enable if you agree.