Kibana observability_ai_assistant API
Interact with the Observability AI Assistant resources.
OpenAPI Specification
openapi: 3.0.3
info:
contact:
name: Kibana Team
description: 'The Kibana REST APIs enable you to manage resources such as connectors, data views, and saved objects.
The API calls are stateless.
Each request that you make happens in isolation from other calls and must include all of the necessary information for Kibana to fulfill the
request.
API requests return JSON output, which is a format that is machine-readable and works well for automation.
To interact with Kibana APIs, use the following operations:
- GET: Fetches the information.
- PATCH: Applies partial modifications to the existing information.
- POST: Adds new information.
- PUT: Updates the existing information.
- DELETE: Removes the information.
You can prepend any Kibana API endpoint with `kbn:` and run the request in **Dev Tools → Console**.
For example:
```
GET kbn:/api/data_views
```
For more information about the console, refer to [Run API requests](https://www.elastic.co/docs/explore-analyze/query-filter/tools/console).
NOTE: Access to internal Kibana API endpoints will be restricted in Kibana version 9.0. Please move any integrations to publicly documented APIs.
## Documentation source and versions
This documentation is derived from the `main` branch of the [kibana](https://github.com/elastic/kibana) repository.
It is provided under license [Attribution-NonCommercial-NoDerivatives 4.0 International](https://creativecommons.org/licenses/by-nc-nd/4.0/).
This documentation contains work-in-progress information for future Elastic Stack releases.
'
title: Kibana APIs Actions observability_ai_assistant API
version: ''
x-doc-license:
name: Attribution-NonCommercial-NoDerivatives 4.0 International
url: https://creativecommons.org/licenses/by-nc-nd/4.0/
x-feedbackLink:
label: Feedback
url: https://github.com/elastic/docs-content/issues/new?assignees=&labels=feedback%2Ccommunity&projects=&template=api-feedback.yaml&title=%5BFeedback%5D%3A+
servers:
- url: https://{kibana_url}
variables:
kibana_url:
default: localhost:5601
security:
- apiKeyAuth: []
- basicAuth: []
tags:
- description: Interact with the Observability AI Assistant resources.
externalDocs:
description: Observability AI Assistant
url: https://www.elastic.co/docs/solutions/observability/observability-ai-assistant
name: observability_ai_assistant
x-displayName: Observability AI Assistant
paths:
/api/observability_ai_assistant/chat/complete:
post:
description: '**Spaces method and path for this operation:**
<div><span class="operation-verb post">post</span> <span class="operation-path">/s/{space_id}/api/observability_ai_assistant/chat/complete</span></div>
Refer to [Spaces](https://www.elastic.co/docs/deploy-manage/manage-spaces) for more information.
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.
'
operationId: observability-ai-assistant-chat-complete
requestBody:
content:
application/json:
examples:
chatCompleteRequestExample:
$ref: '#/components/examples/Observability_AI_Assistant_API_ChatCompleteRequestExample'
schema:
type: object
properties:
actions:
items:
$ref: '#/components/schemas/Observability_AI_Assistant_API_Function'
type: array
connectorId:
description: A unique identifier for the connector.
type: string
conversationId:
description: A unique identifier for the conversation if you are continuing an existing conversation.
type: string
disableFunctions:
description: Flag indicating whether all function calls should be disabled for the conversation. If true, no calls to functions will be made.
type: boolean
instructions:
description: An array of instruction objects, which can be either simple strings or detailed objects.
items:
$ref: '#/components/schemas/Observability_AI_Assistant_API_Instruction'
type: array
messages:
description: An array of message objects containing the conversation history.
items:
$ref: '#/components/schemas/Observability_AI_Assistant_API_Message'
type: array
persist:
description: Indicates whether the conversation should be saved to storage. If true, the conversation will be saved and will be available in Kibana.
type: boolean
title:
description: A title for the conversation.
type: string
required:
- messages
- connectorId
- persist
responses:
'200':
content:
application/json:
examples:
chatCompleteResponseExample:
$ref: '#/components/examples/Observability_AI_Assistant_API_ChatCompleteResponseExample'
schema:
type: object
description: Successful response
summary: Generate a chat completion
tags:
- observability_ai_assistant
x-codeSamples:
- lang: cURL
source: "curl --request POST 'localhost:5601/api/observability_ai_assistant/chat/complete' -u <username>:<password> -H 'kbn-xsrf: true' -H \"Content-Type: application/json\" --data '\n{\n\"connectorId\": \"<connectorId>\",\n\"disableFunctions\": false,\n \"messages\": [\n {\n \"@timestamp\": \"2025-06-25T23:45:00.000Z\",\n \"message\": {\n \"role\": \"user\",\n \"content\": \"Is my Elasticsearch cluster healthy right now?\"\n }\n }\n ],\n\"persist\": false,\n\"actions\": [\n {\n \"name\": \"get_cluster_health\",\n \"description\": \"Fetch the current Elasticsearch cluster-health status and key metrics.\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"includeShardStats\": {\n \"type\": \"boolean\",\n \"default\": false\n }\n }\n }\n }\n],\n\"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.\"]\n}'\n"
x-state: Technical Preview
x-metaTags:
- content: Kibana
name: product_name
components:
examples:
Observability_AI_Assistant_API_ChatCompleteRequestExample:
summary: Example of completing a chat interaction
value: "{\n \"connectorId\": \"<connectorId>\",\n \"disableFunctions\": false,\n \"messages\": [\n {\n \"@timestamp\": \"2025-06-25T23:45:00.000Z\",\n \"message\": {\n \"role\": \"user\",\n \"content\": \"Is my Elasticsearch cluster healthy right now?\"\n }\n }\n ],\n \"persist\": false,\n \"actions\": [\n {\n \"name\": \"get_cluster_health\",\n \"description\": \"Fetch the current Elasticsearch cluster-health status and key metrics.\",\n \"parameters\": {\n \"type\": \"object\",\n \"properties\": {\n \"includeShardStats\": {\n \"type\": \"boolean\",\n \"default\": false\n }\n }\n }\n }\n ],\n \"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.\"]\n}\n"
Observability_AI_Assistant_API_ChatCompleteResponseExample:
summary: Get a chat completion from the Observability AI Assistant
value: '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]
'
schemas:
Observability_AI_Assistant_API_FunctionCall:
description: Details of the function call within the message.
type: object
properties:
arguments:
description: The arguments for the function call.
type: string
name:
description: The name of the function.
type: string
trigger:
description: The trigger of the function call.
enum:
- assistant
- user
- elastic
type: string
required:
- name
- trigger
Observability_AI_Assistant_API_Message:
name: Message
type: object
properties:
'@timestamp':
description: The timestamp when the message was created.
type: string
message:
description: The main content of the message.
type: object
properties:
content:
description: The content of the message.
type: string
data:
description: Additional data associated with the message.
type: string
event:
description: The event related to the message.
type: string
function_call:
$ref: '#/components/schemas/Observability_AI_Assistant_API_FunctionCall'
name:
description: The name associated with the message.
type: string
role:
$ref: '#/components/schemas/Observability_AI_Assistant_API_MessageRoleEnum'
required:
- role
required:
- '@timestamp'
- message
Observability_AI_Assistant_API_MessageRoleEnum:
description: The role of the message sender.
enum:
- system
- assistant
- function
- user
- elastic
type: string
Observability_AI_Assistant_API_Function:
type: object
properties:
description:
description: The description of the function.
type: string
name:
description: The name of the function.
type: string
parameters:
description: The parameters of the function.
type: object
Observability_AI_Assistant_API_Instruction:
oneOf:
- description: A simple instruction represented as a string.
type: string
- description: A detailed instruction with an ID and text.
type: object
properties:
id:
description: A unique identifier for the instruction.
type: string
text:
description: The text of the instruction.
type: string
required:
- id
- text
securitySchemes:
apiKeyAuth:
description: 'These APIs use key-based authentication. You must create an API key and use the encoded value in the request header. For example: `Authorization: ApiKey base64AccessApiKey`
'
in: header
name: Authorization
type: apiKey
basicAuth:
scheme: basic
type: http
x-topics:
- title: Kibana spaces
content: "Spaces enable you to organize your dashboards and other saved objects into meaningful categories.\nYou can use the default space or create your own spaces.\n\nTo run APIs in non-default spaces, you must add `s/{space_id}/` to the path.\nFor example:\n\n```bash\ncurl -X GET \"http://${KIBANA_URL}/s/marketing/api/data_views\" \\\n -H \"Authorization: ApiKey ${API_KEY}\"\n```\n\nIf you use the Kibana console to send API requests, it automatically adds the appropriate space identifier.\n\nTo learn more, check out [Spaces](https://www.elastic.co/docs/deploy-manage/manage-spaces).\n"