Skip to main content
POST
Recall facts by meaning

Authorizations

Authorization
string
header
required

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

Body

application/json

Retrieve facts by meaning. The query is embedded and compared against every fact in the scope.

query
string
required

What to recall. Max 8,192 bytes. Phrase it as the fact you hope to find, not as a keyword: the match is semantic.

Example:

"when is a good time to reach them"

scope
enum<string>

Which namespace a memory belongs to. Scopes are isolated: a search in one never returns a memory from another.

  • project: shared by the whole project. What the agent knows in general.
  • contact: what the agent knows about one person, across every conversation with them.
  • conversation: what the agent knows about one inbox thread.

Defaults to project everywhere. contact and conversation require the matching id.

Available options:
project,
contact,
conversation
contactId
string

Required when scope is contact.

conversationId
string

Required when scope is conversation.

limit
integer
default:5

Maximum hits to return, closest first.

Required range: 1 <= x <= 20
minScore
number

Drop hits below this similarity. Without it every scope returns its limit closest facts, however far away they are — an empty-handed search still answers with something. Set it (0.3 or above is a reasonable floor) when "nothing relevant" must be distinguishable from "the least irrelevant thing I hold".

Required range: 0 <= x <= 1

Response

Matching facts, closest first. An empty items means nothing cleared minScore.

items
object[]
required