See what the camera could actually read
The custom receipt camera runs an on-device text pass after capture. In the review step, recognised lines are highlighted over the image so you can spot a missed total, folded corner or blurry line before continuing.
If no clear text is found, the photo is not silently discarded. You can retake it or continue to an editable result.
- Custom camera and framing
- Recognised text highlighted on the review image
- Recoverable low-confidence captures
- No bank connection required
From one total to the items inside it
A supermarket transaction can hide several kinds of spending. VULT can preserve the transaction total while allocating the receipt items underneath it — for example pantry goods, fresh food or household supplies. You choose whether analytics stay at group, budget-line or product level.
Cloud-assisted matching is optional and runs through a protected server function. It works from bounded extracted text and category context rather than treating an automated result as unquestionable. Every category and amount remains editable.
- Merchant, date and total extraction
- Itemised allocations that still reconcile to the transaction
- Category suggestions based on the budget you already use
- Manual correction before or after saving
Private by default, honest about assistance
Receipt photos start with on-device recognition. Where an eligible cloud match is requested, VULT sends only the information required for that task through a protected backend. Automated interpretation can be wrong, so the app always keeps the review step and does not present categorisation as financial advice.