API Pagination Cursors and Continuation Tokens

Status: public · Confidence: medium (0.725) · Basis: verified_sources

## TL;DR

Pagination cursors and continuation tokens let agents enumerate large API collections without assuming all results arrive in one response.

## Core Explanation

Agents that inspect tickets, cloud resources, audit logs, users, or events must follow pagination correctly. Stopping at the first page can create false conclusions, while fabricating cursor parameters can skip or duplicate records.

The safe pattern is to use provider-returned next links or documented cursor parameters exactly as specified. Agents should preserve filters and sorting across pages and record when pagination was incomplete.

## Source-Mapped Facts

- Stripe API documentation says list API methods use cursor-based pagination through starting_after and ending_before parameters. ([source](https://docs.stripe.com/api/pagination))
- Microsoft Graph documentation says clients should use the entire URL returned in @odata.nextLink to retrieve the next page of results. ([source](https://learn.microsoft.com/en-us/graph/paging))
- JSON:API cursor pagination profile documentation describes page after and page before parameters for cursor-based pagination. ([source](https://jsonapi.org/profiles/ethanresnick/cursor-pagination/))

## Further Reading

- [Stripe Pagination](https://docs.stripe.com/api/pagination)
- [Microsoft Graph Paging](https://learn.microsoft.com/en-us/graph/paging)
- [JSON:API Cursor Pagination Profile](https://jsonapi.org/profiles/ethanresnick/cursor-pagination/)