document.processed

Fires once per document, after its extraction has been stored. Carries routing identifiers only, so read the extracted figures back through the API by document_id.