> ## Documentation Index
> Fetch the complete documentation index at: https://socialquery.dev/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Responses and pagination

> Handle the request wrapper, then parse the selected resource.

Start with the HTTP outcome and `success`. Responses identify the `platform` and `endpoint` and include usage metadata. Successful calls supply `data`; unsuccessful calls supply an `error` object.

Interpret `data` using the selected operation’s contract. A list can expose `items`, `total`, or `next_cursor`, while a detail call returns its resource object. A `total` may count only the current page. Preserve absent or null fields instead of assuming they describe zero-valued metrics.

Continue paging with the original endpoint and query inputs. Treat the returned cursor as an opaque continuation value, not a page number or a token transferable to another source. Paging behavior remains specific to the operation during preview.

When an operation includes a `computed` object, profile engagement and estimated reach remain `null`: audience totals do not establish a reporting window or unique reach. For individual posts, `engagement_rate` is `(likes + comments + shares) / views` only when all four counters are available and valid and views are positive. Missing counters remain unknown, not zero. The ratio can exceed 1 because interactions are not unique people. Estimated reach remains `null`; view counts are not multiplied into reach estimates. These rules apply to fresh responses; replaying an earlier idempotent request preserves its original data.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.