Pagination

Every list endpoint returns the same shape:

{
  "object": "list",
  "data": [{ "id": "pay_1042", "object": "payment", "…": "…" }],
  "has_more": true,
  "next_cursor": "eyJpZCI6MTAzOH0"
}

Parameter	Meaning
`limit`	Page size, 1-100. Default 20.
`cursor`	The `next_cursor` of the previous page. Omit it for the first (newest) page.
`created_gte`, `created_lt`	Optional time window (ISO 8601), where the endpoint supports it.

Lists are ordered newest first. has_more: false and next_cursor: null mark the last page.

Paging through a list

The SDK pages for you. for await fetches the next page when it needs it:

for await (const payment of zp.payments.list({ created_gte: '2026-09-01T00:00:00Z', limit: 100 })) {
  await record(payment);
}

Awaiting a list call returns one page:

const page = await zp.payments.list({ limit: 50 });
console.log(page.data.length, page.has_more, page.next_cursor);

Cursors are opaque. Pass them back unchanged. They are tied to one list: a cursor from another list, or a modified one, answers 400 cursor_invalid.
Keep your filters. Send the same filters with every page of one walk.
Stable under inserts. Cursors are keyset-based, not offsets. New objects created while you page appear at the top of the list, never as duplicates or gaps in the pages you are reading.
Time windows for reconciliation. Use created_gte and created_lt to walk one closed period, for example one day in UTC. See Reconciling payments.

GET /v1/webhook_endpoints returns all endpoints of the mode (at most 16) in one list.

Events are kept for 30 days: GET /v1/events never goes further back.