Docs
OverviewGuidesZero & MCPAPI referenceChangelog
Knowledge Base

List Documents

API key access

Needs the kb:read scope with an API key (ro_pat_...). A signed-in session uses the caller's role.
GET/api/v1/kb/documents

Authorization

AuthorizationBearer <token>

Access token from /api/v1/auth/login, or an API key (ro_pat_..., RO-176) from /api/v1/workspaces/current/api-keys: the key implies its workspace and needs the route's scope. MCP connectors (/api/v1/mcp only) may use an OAuth access token (ro_oat_..., RO-284).

In: header

Query Parameters

limit?Limit
Default50
Range1 <= value <= 200
offset?Offset
Default0
Range0 <= value

Header Parameters

X-Workspace-Id?X-Workspace-Id

Id of the workspace to act on (must be a member).

Response Body

curl -X GET "http://localhost:8000/api/v1/kb/documents?limit=50&offset=0" \  -H "X-Workspace-Id: string"
{
  "items": [
    {
      "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08",
      "source_type": "url",
      "source_url": "string",
      "title": "string",
      "content_type": "string",
      "status": "pending",
      "error": "string",
      "char_count": 0,
      "chunk_count": 0,
      "embedding_model": "string",
      "created_at": "2019-08-24T14:15:22Z",
      "updated_at": "2019-08-24T14:15:22Z"
    }
  ],
  "total": 0
}
{
  "detail": [
    {
      "loc": [
        "string"
      ],
      "msg": "string",
      "type": "string",
      "input": null,
      "ctx": {}
    }
  ]
}

Ingest Url

Queue a website crawl (same site, robots.txt honored, page-limited) into the knowledge base. The URL is checked now (422 for an invalid, private or unresolvable address); the crawl and embedding run as a background job (RO-068). Poll the returned document with `GET /kb/documents/{id}` until it is `ready` or `failed`. A second request for a URL whose crawl is still pending or running returns that crawl.

Search

Semantic search over the workspace's knowledge base (cosine similarity, best first).