Ingest rows into a live-dataset table

POST

/api/ingest/{collectionName}/{tableName}

curl --request POST \
  --url https://firma.company-gpt.com/companygpt/rag/api/api/ingest/example/example \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{ "rows": [ {} ] }'

• CompanyGPT instance

Insert rows into a “live” dataset table. Requires admin (write) access to the collection. Rows are validated and coerced against the table’s schema; invalid rows are rejected individually rather than failing the whole request. Capped at 10,000 rows and 64 MiB per request.

Authorizations

• BearerAuth

Parameters

Path Parameters

collectionName

required

string

Collection’s technical name

tableName

required

string

Request Body^required

Media typeapplication/json

object

rows

required

Array<object>

<= 10000 items

object

key

additional properties

any

Example^generated

{
  "rows": [
    {}
  ]
}

Responses

200

Empty rows array — nothing to insert

Media typeapplication/json

object

inserted

required

integer

rejected

required

integer

errors

required

Array<object>

object

index

required

0-based index into the request’s “rows” array

integer

error

required

string

Example^generated

{
  "inserted": 1,
  "rejected": 1,
  "errors": [
    {
      "index": 1,
      "error": "example"
    }
  ]
}

201

Rows processed (individual row failures are reported in “errors”, not as an HTTP error)

Media typeapplication/json

object

inserted

required

integer

rejected

required

integer

errors

required

Array<object>

object

index

required

0-based index into the request’s “rows” array

integer

error

required

string

Example^generated

{
  "inserted": 1,
  "rejected": 1,
  "errors": [
    {
      "index": 1,
      "error": "example"
    }
  ]
}

400

Invalid table name, malformed body, or more than 10,000 rows

401

Unauthorized

403

Admin access required to ingest data

404

Dataset (or table) not found — also returned for any collection that is not a live dataset