AI and memory

agents.playground.receipt

Dry run of the receipt reader: read a sample receipt (filename, mimeType, contentBase64 of up to 2 MB; JPEG, PNG, WebP, HEIC, PDF or .eml) the same way intake does and return the extracted facts and validation flags. Nothing is saved: no submission, evidence, entry or run.

GETPOST/api/v1/accounting/agents.playground.receipt

Permissions

accounting:read

Who can call it

admin, finance

Retries

Safe to retry: it only reads

MCP

accounting_agents_playground_receipt

Fields

  • filenamestringRequired

    The file's name, including its extension.

    1–200 characters

  • mimeTypeenumRequired

    The file's type, such as image/jpeg or application/pdf.

    image/jpegimage/pngimage/webpimage/heicimage/heifapplication/pdfmessage/rfc822
  • contentBase64stringRequired

    The file's exact bytes, encoded as base64.

    4–2800000 characters · Matches ^[A-Za-z0-9+/]+={0,2}$

Example

curl -G https://app.getoatmilk.com/api/v1/accounting/agents.playground.receipt \
  -H "Authorization: Bearer $OATMILK_API_KEY" \
  --data-urlencode 'filename=receipt.jpg' \
  --data-urlencode 'mimeType=image/jpeg' \
  --data-urlencode 'contentBase64=contentBase64'
Response
{
  "data": { … }
}

Try it

Try it

Checks your input with this action’s real schema and answers like the API, with synthetic data. No key needed, and nothing changes.

GET/api/v1/accounting/agents.playground.receipt
curl -G https://app.getoatmilk.com/api/v1/accounting/agents.playground.receipt \
  -H "Authorization: Bearer $OATMILK_API_KEY" \
  --data-urlencode 'filename=receipt.jpg' \
  --data-urlencode 'mimeType=image/jpeg' \
  --data-urlencode 'contentBase64=contentBase64'

More in AI and memory.