Omni Docs
API Reference

Memories & Semantic Search

Endpoints for managing persistent facts, executing vector similarity queries, and building context bundles.

POST /v1/memories

Stores a new semantic memory into the persistent memory store.

  • Scope Required: memory:write

Request Body

{
  "content": "User prefers concise Python code examples.",
  "metadata": {
    "category": "preferences",
    "source": "developer-settings"
  }
}

Response (200 OK)

{
  "id": "mem_01jze9m2xw",
  "ownerId": "usr_94a645d99c3a",
  "content": "User prefers concise Python code examples.",
  "metadata": { "category": "preferences" },
  "createdAt": "2026-08-16T00:00:00.000Z",
  "updatedAt": "2026-08-16T00:00:00.000Z"
}

GET /v1/memories

Lists all memories belonging to the authenticated account.

  • Scope Required: memory:read

POST /v1/memories/search

Searches memories using vector semantic similarity.

  • Scope Required: memory:read

Request Body

{
  "query": "coding preferences",
  "limit": 5
}

Response

{
  "results": [
    {
      "id": "mem_01jze9m2xw",
      "content": "User prefers concise Python code examples.",
      "score": 0.89,
      "createdAt": "2026-08-16T00:00:00.000Z"
    }
  ],
  "mode": "semantic"
}

POST /v1/context/build

Builds an integrated context block containing matched memories and workspace notes.

  • Scope Required: memory:read

Request Body

{
  "query": "NDA expiration term",
  "workspace_id": "ws_123",
  "limit": 5,
  "include": {
    "memories": true,
    "notes": true,
    "files": false
  }
}

Response

{
  "query": "NDA expiration term",
  "mode": "semantic",
  "items": [
    {
      "type": "note",
      "id": "doc_123",
      "score": 0.92,
      "content": "Mutual confidentiality term is 3 years."
    }
  ],
  "contextText": "Relevant context:\n- Mutual confidentiality term is 3 years."
}

On this page