Pages crawled

curl --request GET \
  --url 'https://truestat.io/api/v1/sites/{siteId}/ai/pages' \
  --header 'Authorization: Bearer <token>'

{
  "data": [
    {
      "path": "/",
      "views": 703,
      "purpose": "ai_training"
    },
    {
      "path": "/pricing",
      "views": 88,
      "purpose": "search"
    }
  ],
  "meta": {
    "page": {
      "limit": 100,
      "offset": 0,
      "has_more": true
    },
    "range": {
      "to": "2026-09-16T13:40:00.000Z",
      "tz": "UTC",
      "from": "2026-09-09T13:40:00.000Z",
      "clamped": false
    },
    "site_id": "6fed8980-a119-4a9c-a4d4-53e448ef9f9e"
  }
}

A ranked list of the pages crawlers actually fetched.

Fetched is not the same as cited

This endpoint and referrals look similar and count opposite things:

The interesting reading is the gap. A page crawled heavily and never referred is being consumed rather than recommended — read into a model, not handed to a reader. A page with few fetches and steady referrals is the opposite: it gets cited, and citations bring people.

Neither is automatically good or bad. But they are different things, and a tool that adds them together tells you nothing about either.

purpose travels with the page

Each row carries the purpose of the fetch, so the same path appears more than once when it was fetched for more than one reason. /pricing under search and /pricing under ai_training are two rows, and it usually matters which one grew.

Paging

The default page is 100 rows, ordered by fetches. limit and offset work as they do everywhere else, and meta.page.has_more tells you whether there is another page — it is measured, not guessed from a row count.

Authorizations

Path parameters

Query parameters

Response