Receipt scanner

Read a selected receipt locally and ask the owner to confirm extracted text.

Documentation revision 1 · Contract reference API 54 · Installed readiness unknown

Preview offline card · Complete HTML interaction body · Machine-readable recipe

What runs where

Owner-selected gallery image → on-device Latin OCR → controller validates totals/date/currency → HTML confirmation.
OCR is local. The card can run OCR through Bridgit.act while the controller is offline, but categorizing, deduplicating and storing a ledger need the controller.

Controller integration

Select the authorized gallery_id, populate the card configuration, save only an owner-confirmed answer.
Controller owns durable receipt storage and deduplication; respond acknowledges a request, not a saved ledger entry.

Only explicit decision-model/weather requests leave the phone; direct or relay transport still requires a running reachable phone.

Boundaries

OCR text is not a verified accounting record. This recipe does not pay, submit expenses, or infer a currency.
gallery_id 42 is a synthetic fixture: substitute the image ID returned by the owner-authorized gallery listing. No camera capture occurs on opening the template.

Lifecycle

start

Configure card data, POST the generated HTML interaction; this does not automatically capture or poll.

update

Owner explicitly refreshes; controller stores approved data and posts a replacement card as needed, cancelling old pending cards.

pause

The Pause button suspends this card’s controls and suppresses results from a pending refresh. There is no background job in this starter; an in-flight capture may finish. Controller integrations must separately pause their jobs.

stop

The Stop button stops this card, suppresses late results, and requests controller shutdown through one response. Confirm/close the controller job and DELETE the interaction separately. Local stop is not proof that an external process stopped.

Failures and costs

No image/grant/model access: show error, retain previous draft with its timestamp.
Blurred/empty text: request another owner-selected image; never invent a missing amount.
OCR has no paid remote inference. An optional controller model has its own costs and privacy boundary.

Complete synthetic contract examples

These fixtures demonstrate contracts. They are not observations from a phone. Replace image/resource IDs with returned authorized IDs. Load request.json as the complete body of POST /v1/interactions; it contains the full offline HTML.

read

{
  "id": "read",
  "request": {
    "method": "POST",
    "path": "/v1/ocr",
    "body": {
      "gallery_id": 42
    }
  },
  "response": {
    "status": 200,
    "body": {
      "text": "CORNER SHOP\n2026-11-04\nTOTAL 12.50 EUR",
      "script": "Latin",
      "on_device": true
    }
  },
  "evidence": "Synthetic contract fixture, not a device observation."
}

Extensions

Add barcode recognition to match products.
Capture a receipt with POST /v1/camera/snapshot, then pass its returned data_base64 to /v1/ocr.