SmartMemory
POST
/memory/ontology/inference/run

Run ontology inference over provided raw text chunks.

Requires authentication and operates on user-scoped data. Returns a JSON-serializable payload with changeset (OntologyIR as dict), metrics, and status.

Request Body

application/json

TypeScript Definitions

Use the request body type in TypeScript.

Response Body

application/json

application/json

curl -X POST "https://example.com/memory/ontology/inference/run" \  -H "Content-Type: application/json" \  -d '{}'
null
{
  "detail": [
    {
      "loc": [
        "string"
      ],
      "msg": "string",
      "type": "string",
      "input": null,
      "ctx": {}
    }
  ]
}

Run Inference

Run inference engine to create inferred edges from rules.

Run Sleep Cycle Now

Fire one sleep cycle synchronously for the caller's workspace. Bypasses the schedule but NOT the cross-replica lock — a concurrent run for the same workspace returns 409 ``lock_held`` on the loser (mirrors ``summary.generate_snapshot``).