List searchable fields
/resources/fieldsReturns the searchable fields for this instance, resolved live from
the box's resource_type_field configuration. The field set is
instance-specific — each publisher/deployment exposes different
fields — so an integration should discover fields at runtime rather than
assume a fixed set. The name of each field is what you pass as a key in
fieldFilters/dateFilters on POST /resources/search.
Authorization
apiKey Issued to you when your integration is provisioned. Determines which catalogue resources are visible and which AI tools are available.
In: header
Query Parameters
Optionally restrict to a resource-type's fields.
Response Body
application/json
curl -X GET "https://example.com/resources/fields"{ "data": { "fields": [ { "ref": 0, "name": "string", "title": "string", "type": 0 } ] }}Tree/category field options GET
For tree- or list-typed fields, returns the available node options.
Get a resource's extracted full text (paginated) GET
Returns a resource's extracted full text, paginated by character. This is the direct-REST door for the AI chat's `document_read` tool (full content) and reads the same extracted-text source — a big enabler for external RAG, since a caller can pull the text and chunk/embed it themselves. Page with either `offset` (0-based character offset) or `page` (1-based; `offset = (page - 1) * limit`). When both are present, `offset` wins. `total` is the full extracted-text length; `hasMore` is true while further pages remain. A resource that exists and is visible but has no extracted text yet returns `200` with `total: 0` and empty `content` — not a 404. Not-found and no-view-access both return `404 NOT_FOUND` so the surface never reveals which refs exist in other usergroups.