Get the intelligence context
Everything this workspace knows, compiled and bounded, in one read. It returns every active document in a fixed order, capped at 18,000 characters each and 72,000 in total. A document cut to fit says so with `truncated`. Free, and there is no quote: the compiler runs no model.
Everything this workspace knows, compiled and bounded, in one read.
It returns every active document in a fixed order, capped at 18,000 characters each and 72,000 in total. A document cut to fit says so with truncated.
Free, and there is no quote: the compiler runs no model.
Authorization
bearerAuth API key authentication. Use your API key as the bearer token: Authorization: Bearer roasos_sk_...
In: header
Response Body
application/json
application/json
application/json
curl -X GET "https://example.com/v1/intelligence/context"{ "data": { "schema_version": "intelligence-context@1", "kind": "intelligence_context", "documents": [ { "slug": "brand-truth", "title": "string", "version": 1, "updated_at": "2019-08-24T14:15:22Z", "markdown": "string", "truncated": true } ], "total_characters": 0, "max_characters_per_document": 18000, "max_total_characters": 72000 }}Replace the marketing brief
Replace the stored marketing brief. Send confirmed values only, in your customers' own words: the words-to-use and the proof points anchor every script the agent writes. This door REPLACES the brief, so send the whole thing; the in-app agent merge-patches instead, which is why the two behave differently. Send at least one field. Free, and there is no quote.
List intelligence documents
The compiled documents this workspace has: the slug, the title, the version, a one-line summary and when each was last compiled. A slug with no source data has no document at all. An absent document is the honest answer; roasOS does not return a stub that says "learning". Free, and there is no quote.