Pagination and large results
Reports are single objects, so most calls need no paging. Two places return long lists.
Request history
GET /v1/requests lists calls on your wallet, newest first, across every key.
| Parameter | Default | Range |
|---|---|---|
limit | 20 | 1–100 |
before | none | a next_cursor value from the previous page |
curl -s "https://getallapis.com/v1/requests?limit=50" -H "Authorization: Bearer $GETALLAPIS_API_KEY"
# then
curl -s "https://getallapis.com/v1/requests?limit=50&before=<next_cursor>" -H "Authorization: Bearer $GETALLAPIS_API_KEY"
Keep requesting with before=<next_cursor> until next_cursor is null. Cursors are stable: new calls made while you page do not shift the pages you have not read yet.
Long lists inside a report
An audit can list many findings, URLs and gaps. Instead of paging, shape the response:
max_items=10caps every list inoutput;_truncatedsays how many rows were withheld per path.fields=summary,actionskeeps only the parts you need.summary=truereturns an outline of the output (types, counts, sizes) so an agent can decide what to fetch.
Shaping never changes the price. The full result stays stored for your retention window and can be re-read for free with GET /v1/results/{request_id}, with different controls each time. See Running endpoints.
Keyword gap size
seo.keyword_gap takes limit (default 25, maximum 100) for the number of gap rows returned. The price is per report, not per row.
Plain Markdown for agents: /docs/pagination.md · index: llms.txt