Privacy

Document content: line items, prices, client names, addresses, tax identifiers, bank details and notes. These exist only in your browser tab and, if you press Save, in IndexedDB on this device.

There is no account system, so there is no profile to attach data to. There is no server-side PDF renderer, so your document is never posted to an API route for processing.

IndexedDB database OpenDocuHub_Vault: saved documents, clients, catalogue items, document sequences and seller settings.

localStorage: an unsaved draft per tool, so a refresh does not destroy work, and a dismissal flag for the draft banner.

Cache Storage: the app shell and engine chunks used by the service worker for offline use.

All of it can be exported as JSON and deleted from the Vault panel. Wiping site data removes it completely.

A share link compresses the document state into the URL fragment after the #. Browsers do not transmit fragments to servers. Anyone holding the link can read the payload, so treat it like the document itself.

Hosting serves static and server-rendered marketing pages and may keep standard request logs (IP address, user agent, timestamp). Those logs do not contain document data.

Receipt OCR downloads an English language model from tessdata.projectnaptha.com the first time you scan an image. That request fetches a public model file; the image itself is processed locally and is never uploaded.

Because document data is not held by us, access, portability, rectification and erasure are performed by you, in the Vault panel. For questions about the marketing site or hosting logs, write to hello@opendocuhub.com.