Search the invoices of the organization, newest first, by cursor
Newest first: by issue_date (the Lima wall clock that is printed), then by id, read by cursor (pass next_cursor as after; has_more says whether there is more). The filter is the same as GET /v1/invoices/export, so the CSV holds what this shows. q is read by the API and interpreted says how it was understood (type, series, number, customer_id_type, customer_id_number, customer_name, customers_matched, customers_truncated); a name that matches more than 200 customers uses the 200 most recently invoiced and says customers_truncated: true. CHANGED (2026-11, beta): a row is LIGHT. The listing is an index and the detail the truth: each row carries id, object, environment, full_number, type, series, number, issue_date, customer (id, id_type, id_number, name), establishment, currency, totals.total, status, status_reason, settled, held and void ({ status } of its latest void, or null; a void on its way leaves status as accepted). Items, files, emails, SUNAT and template are only in GET /v1/invoices/{id}.
https://api.invox.page/v1/invoicesAuthorization
Query parameters
- qoptionalstring
One search box, read by the API (
interpretedin the answer says how):F001-123(orF001-00000123,F001 123) a series and number;F001alone a series of yours; 11 digits the customer's RUC; 8 digits a DNI; another word with digits, from 6 to 20 characters, a document of any type (a shorter one is part of the name:caja 1);factura(s)orboleta(s)the type; anything else the customer's name (contains, without case or accents). The words combine with AND:boleta juan perez. Searching by name needsinvoices:read, notcustomers:read.maxLength: 100
Header parameters
Returns
200
A page of light rows, has_more, next_cursor and interpreted.