Check Verification Document Cautions
Header Parameters
Request Body
application/json
TypeScript Definitions
Use the request body type in TypeScript.
Body of POST /verifications/document-cautions.
Response Body
application/json
application/json
curl -X POST "https://example.com/v1/verifications/document-cautions" \ -H "Content-Type: application/json" \ -d '{ "quote_id": "3c071a1d-db86-46a7-9dc8-72ba3fbca992", "country_code": "st", "number_type": "string", "subject_type": "person" }'{ "warnings": [ { "field": "string", "kind": "unreadable", "message": "string" } ]}Check Domain
Does this domain already receive mail? Drives apex-vs-prefix onboarding.
Check Verification Documents
Read the documents of a number verification form before you submit it (multipart/form-data). Send the documents first: the same parts as `POST /verifications`, with a file per slot to read (`file.<slot>`, or `upload.<slot>` with the id of a `verification_document` upload). Only the slots with a file in this request are read. A file that stands for other slots too (one passport for several proofs) is sent once, for its slot, with `documents.<slot>.covers` listing the other slots: it is read once and fills all of them. One file given for several slots uses one read. A `file.<slot>` part for a slot not in `documents` is not read and comes back `read` false. `quote_id`, `country_code`, `number_type` and `subject_type` are required. `documents` may leave out `option` while the type is not chosen; `fields` and `address` may be empty or partial. Returns the values the files show, by the carrier's field names, to fill the form (`fields`, `address`, `documents.<slot>.fields`, and the document type each file looks like in `documents.<slot>.option`). Also returns `warnings`: an unreadable, upside-down or expired document, a file that does not match the chosen document type, or a typed name that differs from the printed one. Show each warning with its fix. You can still submit. A file nothing was read from has `read` false: with an `unreadable` warning when the file is not clear, else with no warning (the read failed or took too long, the file is over the size or page limits, or the workspace used its 30 reads for today). Nothing is filed, saved, held or charged, and uploads are not used up. The files are read by an Anthropic AI model, on AWS (Amazon Bedrock) or Anthropic's API, for this check only.