Read limits from the schema
Each file field can set its own types and size. Retrieve the application schema (GET /v1/onboarding/partner/applications/{applicationId}/schema) and read
x-fileUpload on the target field:
x-fileUpload is absent, the API uses the defaults in the following tables.
Filename
The API keeps the last path component offileName when it contains / or \,
then trims spaces at each end before validating the remaining name. For example,
folder/passport.pdf is stored as passport.pdf.
Commas, quotes, and other punctuation outside the allowlist fail the character
check if they remain in the filename.
File type
The API reads magic bytes infileContent. It doesn’t trust the Content-Type
header.
Known signatures are PDF, JPEG, PNG, ZIP (Office Open XML), and OLE2 (legacy
Office). A file with no matching signature returns unrecognized_type.
Office files (DOC, XLS, DOCX, XLSX) pass only when the field schema lists those
types. The default list doesn’t include them.
File size
A file over the limit returns
too_large.
Error codes
The following checks normally return HTTP400 with numeric code 181102. The
body includes errors with field and code.