Unispec Docs
ApiTyped memory

LLM-extract a record from a conversation or free text

Extracts only explicitly stated values (required is relaxed for the extraction call, then re-enforced before storing). When nothing usable is found the response is stored:false — a normal outcome, not an error. Stored records embed and meter like direct writes.

POST
/v1/records/{schemaName}/extract

Extracts only explicitly stated values (required is relaxed for the extraction call, then re-enforced before storing). When nothing usable is found the response is stored:false — a normal outcome, not an error. Stored records embed and meter like direct writes.

Authorization

TokenAuth
Authorization<token>

Project API key as Token lb-sk_...

In: header

Path Parameters

schemaName*string

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/v1/records/string/extract" \  -H "Content-Type: application/json" \  -d '{    "text": "Hi, I am Ada — email ada@example.com. Hoping to buy this quarter.",    "user_id": "visitor-42"  }'
{  "stored": true,  "record": {}}
{  "stored": true,  "record": {    "id": "string",    "schema": "string",    "schema_version": 0,    "user_id": "string",    "agent_id": "string",    "run_id": "string",    "app_id": "string",    "record": {},    "created_at": "2019-08-24T14:15:22Z"  }}