Create a model response | Kibana Serverless API

Create a model response Beta

POST /api/security_ai_assistant/chat/complete

Create a model response for the given chat conversation.

application/json; Elastic-Api-Version=2023-10-31

Body Required

connectorId string Required
conversationId string
isStream boolean
langSmithApiKey string
langSmithProject string
messages array[object] Required
Hide messages attributes Show messages attributes object
- content string
  
  Message content.
- data object
  
  ECS object to attach to the context of the message.
  
  Additional properties are allowed.
- fields_to_anonymize array[string]
- role string Required
  
  Message role.
  
  Values are system, user, or assistant.
model string
persist boolean Required
promptId string
responseLanguage string

Responses

200 application/octet-stream; Elastic-Api-Version=2023-10-31

Indicates a successful call.
400 application/json; Elastic-Api-Version=2023-10-31

Generic Error
Hide response attributes Show response attributes object
- error string
- message string
- statusCode number

POST /api/security_ai_assistant/chat/complete

curl \
 -X POST https://localhost:5601/api/security_ai_assistant/chat/complete \
 -H "Content-Type: application/json; Elastic-Api-Version=2023-10-31"

Request examples

{
  "connectorId": "string",
  "conversationId": "string",
  "isStream": true,
  "langSmithApiKey": "string",
  "langSmithProject": "string",
  "messages": [
    {
      "content": "string",
      "data": {},
      "fields_to_anonymize": [
        "string"
      ],
      "role": "system"
    }
  ],
  "model": "string",
  "persist": true,
  "promptId": "string",
  "responseLanguage": "string"
}

Response examples (400)

{
  "error": "string",
  "message": "string",
  "statusCode": 42.0
}