Generate a chat completion | Kibana API documentation

Generate a chat completion Technical Preview

POST /api/observability_ai_assistant/chat/complete

Create a new chat completion by using the Observability AI Assistant.

The API returns the model's response based on the current conversation context.

It also handles any tool requests within the conversation, which may trigger multiple calls to the underlying large language model (LLM).

This functionality is in technical preview and may be changed or removed in a future release. Elastic will work to fix any issues, but features in technical preview are not subject to the support SLA of official GA features.

application/json

Body

actions array[object]
Hide actions attributes Show actions attributes object
- description string
  
  The description of the function.
- name string
  
  The name of the function.
- parameters object
  
  The parameters of the function.
connectorId string Required

A unique identifier for the connector.
conversationId string

A unique identifier for the conversation if you are continuing an existing conversation.
disableFunctions boolean

Flag indicating whether all function calls should be disabled for the conversation. If true, no calls to functions will be made.
instructions array[string | object]

An array of instruction objects, which can be either simple strings or detailed objects.
One of:
string-1 string object-2 object

A simple instruction represented as a string.
A detailed instruction with an ID and text.
Hide attributes Show attributes

id string Required

A unique identifier for the instruction.

text string Required

The text of the instruction.
messages array[object] Required

An array of message objects containing the conversation history.
Hide messages attributes Show messages attributes object
- @timestamp string Required
  
  The timestamp when the message was created.
- message object Required
  
  The main content of the message.
  Hide message attributes Show message attributes object
  
  content string
  
  The content of the message.
  
  data string
  
  Additional data associated with the message.
  
  event string
  
  The event related to the message.
  
  function_call object
  
  Details of the function call within the message.
  
  Hide function_call attributes Show function_call attributes object
  
  arguments string
  
  The arguments for the function call.
  
  name string Required
  
  The name of the function.
  
  trigger string Required
  
  The trigger of the function call.
  
  Values are assistant, user, or elastic.
  
  name string
  
  The name associated with the message.
  
  role string Required
  
  The role of the message sender.
  
  Values are system, assistant, function, user, or elastic.
persist boolean Required

Indicates whether the conversation should be saved to storage. If true, the conversation will be saved and will be available in Kibana.
title string

A title for the conversation.

Responses

200 application/json

Successful response

POST /api/observability_ai_assistant/chat/complete

curl --request POST 'localhost:5601/api/observability_ai_assistant/chat/complete' -u <username>:<password> -H 'kbn-xsrf: true' -H "Content-Type: application/json" --data '
{
"connectorId": "<connectorId>",
"disableFunctions": false,
  "messages": [
    {
      "@timestamp": "2025-06-25T23:45:00.000Z",
      "message": {
        "role": "user",
        "content": "Is my Elasticsearch cluster healthy right now?"
      }
    }
  ],
"persist": false,
"actions": [
  {
    "name": "get_cluster_health",
    "description": "Fetch the current Elasticsearch cluster-health status and key metrics.",
    "parameters": {
      "type": "object",
      "properties": {
        "includeShardStats": {
          "type": "boolean",
          "default": false
        }
      }
    }
  }
],
"instructions": ["When the user asks about Elasticsearch cluster health, use the get_cluster_health tool to retrieve cluster health, then summarize the response in plain English."]
}'

Request example

{
  "connectorId": "<connectorId>",
  "disableFunctions": false,
  "messages": [
    {
      "@timestamp": "2025-06-25T23:45:00.000Z",
      "message": {
        "role": "user",
        "content": "Is my Elasticsearch cluster healthy right now?"
      }
    }
  ],
  "persist": false,
  "actions": [
    {
      "name": "get_cluster_health",
      "description": "Fetch the current Elasticsearch cluster-health status and key metrics.",
      "parameters": {
        "type": "object",
        "properties": {
          "includeShardStats": {
            "type": "boolean",
            "default": false
          }
        }
      }
    }
  ],
  "instructions": ["When the user asks about Elasticsearch cluster health, use the get_cluster_health tool to retrieve cluster health, then summarize the response in plain English."]
}

Response examples (200)

data: {"model":"unknown","choices":[{"delta":{"content":"","function_call":{"name":"get_cluster_health","arguments":"{\"includeShardStats\":true}"}},"finish_reason":null,"index":0}],"created":1750936626911,"id":"9c8eff9b-4fd4-4203-a4ab-2e364688deff","object":"chat.completion.chunk"}

data: [DONE]