> ## Documentation Index
> Fetch the complete documentation index at: https://docs.enneo.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# Get intent(s) detected/assigned for a ticket
## OpenAPI
````yaml https://dev.enneo.dev/api/mind/docs/open-api get /intent/byTicketId/{ticketId}
openapi: 3.0.0
info:
version: '1'
title: enneo.MIND API
description: This describes the API of enneo Mind, the main ticketing backend
contact:
name: enneo GmbH
email: richard@enneo.ai
license:
name: Proprietary software
url: https://enneo.ai
servers:
- url: https://demo.enneo.ai/api/mind
description: Production server, demo client
- url: https://main.enneo.dev/api/mind
description: Development main branch
- url: http://localhost:8005/api/mind
description: Local development server
security:
- bearerAuth:
- api
- cookieAuth:
- api
paths:
/intent/byTicketId/{ticketId}:
get:
tags:
- Intent
summary: Get intent(s) detected/assigned for a ticket
operationId: getIntentByTicketId
parameters:
- name: ticketId
in: path
required: true
description: The id of the ticket for which the intents should be retrieved
schema:
type: integer
example: 376189
responses:
'200':
description: Successful operation
content:
application/json:
schema:
type: object
properties:
ticketId:
type: integer
example: 376189
intentsFound:
description: Number of intents found
type: integer
example: 1
aiOutcome:
description: >
Outcome of the AI's detected processing 1) noDetection: AI
did not find anything useful 2) customerDetected: AI
detected the customer, but nothing else 3) agentAssist: AI
detected an intent, but is not confident enough to
excecute it without additional support 4) full: AI fully
resolved the issue. No agent interaction needed any more
type: string
enum:
- noDetection
- customerDetected
- agentAssist
- full
example: agentAssist
intents:
type: array
items:
$ref: '#/components/schemas/Intent'
'403':
description: Unauthorized
'404':
description: Customer not found
'500':
description: Internal error
content:
application/json:
schema:
$ref: '#/components/schemas/Error'
components:
schemas:
Intent:
type: object
description: An intent of a customer that he wants to be resolved by contacting us
properties:
id:
type: integer
description: Internal ID of intent
example: 1211221
code:
type: string
nullable: true
description: >-
Intent model code that resolves a specific issue. Currently
supported is process_meter_reading, process_bank_data and
process_installment_change
example: process_meter_reading
name:
type: string
description: User-readable name of intent as defined by client
example: Process a meter reading
contractId:
type: string
example: '746839'
status:
type: string
description: >
Lifecycle status of the intent. `preview` — created as a
non-persisted preview. `ready` — active, waiting for agent action.
`executed` — action taken, but dialogue may continue (interaction
with remaining options). `archived` — terminal: action taken and
dialogue is closed (text reply sent, or interaction with no
remaining options). Preserved as audit history. `invalidated` —
underlying data changed, will be re-processed on next load.
`deleted` — soft-deleted.
enum:
- preview
- ready
- executed
- archived
- invalidated
- deleted
example: ready
confidence:
type: number
format: float
description: >-
Confidence how certain the AI is that this is correct. 0.95 means
for example 95% certain. Defaults to 100% when verified by human.
example: 0.95
confidenceColor:
type: string
example: warning
description: >-
A color indication of an intent showing the user how much user
interaction is needed. 'success' indicates the AI is ready to
execute, all other codes require user interaction first
enum:
- success
- neutral
- warning
- danger
verified:
type: boolean
description: True if intent has been verified by human
example: false
context:
type: object
description: >-
Intent-specific additional data to be shown to the user so he has
relevant information about the intent. Data type varies by intent
code
messagePreview:
type: string
nullable: true
description: >-
A preview of the response a customer would receive when executing
this intent. Not always available, e.g. when the respond depends on
user-input
example: We have successfully processed your meter reading
recipient:
type: string
example: john@smith.com
tags:
type: string
description: >-
Tags to show to the user. prePopulated means that all properties are
already pre-filled by the AI and the user can just insert the
template to the body. adaptionNeeded is the opposite, and means the
user still needs to modify the text. writesToErp means that upon
execution of the intent a write is done to the ERP system, like a
insertion of a meter reading or the creation of a bill.
enum:
- prePopulated
- adaptionNeeded
- writesToErp
data:
type: object
description: >-
Intent-specific data object showing all the data that was extracted
from the ticket in a structured form. Data type varies by intent
code
options:
description: >-
Different options that the user is given. Every option is a new
button in the UI
type: array
items:
$ref: '#/components/schemas/IntentOption'
example:
- type: enter_into_system
name: Eintragen
icon: check
recommended: true
order: 1
- type: ignore
name: Ignorieren
icon: cancel
recommended: false
order: 2
- type: forward_to_vnb
name: An VNB verweisen
icon: questionMark
recommended: false
order: 3
infos:
type: array
items:
$ref: '#/components/schemas/IntentInfo'
extraInfo:
type: string
example: null
nullable: true
description: Extra-Information shown in tooltip. Info-Icon only shown if not null
outcome:
$ref: '#/components/schemas/IntentOutcome'
Error:
type: object
description: Data format of Enneo error messages
properties:
error:
type: string
example: Contract 121 could not be processed
description: Readable error message that should be shown to the user
details:
type: string
example: >-
Uncatched null point exception in testFunction() in
/app/src/file:212
description: Not easily readable error message that is for the developer
txId:
type: string
example: c916167c94
description: >-
Internal transaction id. Useful for debugging. Corresponds to the
OpenTelemetry trace ID.
IntentOption:
type: object
properties:
type:
type: string
name:
type: string
icon:
type: string
recommended:
type: boolean
order:
description: Buttons ordered from lowest (shown first) to highest (shown last)
type: integer
IntentInfo:
type: object
description: Info areas shown to the user
properties:
code:
type: string
nullable: true
description: >-
Machine-readable code identifying the info type, e.g. "plausible",
"USER_WARNING", "TRACE_INFO", "ERROR_MESSAGE"
example: USER_WARNING
type:
type: string
example: warning
enum:
- success
- neutral
- warning
- danger
message:
type: string
example: Reading is plausible
extraInfo:
type: string
example: >-
Expected reading was 421 kWh. Plausbible because difference to 317
kWh is below threshold of 200 kWh
description: Extra-Information shown in tooltip. Info-Icon only shown if not null
nullable: true
IntentOutcome:
type: object
nullable: true
description: Shows the outcome (if the intent was already executed)
properties:
aiAgentId:
type: integer
format: int32
description: ID of the AI agent that was executed
example: 1
success:
type: boolean
example: true
messageLocalized:
description: >-
Human-readable message, usually from the backend system. Already
localized.
type: string
example: Powercloud accepted meter reading
internalData:
type: object
description: Technical output for debugging purposes
additionalProperties: true
example:
requestEndpoint: saveReadingByContractId
requestParams: readingValue=21;date=2022-12-31
executedAt:
type: string
description: Date and time when intent was executed
format: DateTime
example: '2022-12-13 22:18:06'
userId:
type: integer
description: User ID of user that triggered the intent. 0 for system user
example: 1
sent:
description: >-
Was the email / chat message / etc. described in this object
successfully sent?
type: boolean
ticketClosed:
description: Was the ticket closed?
type: boolean
example: true
recipient:
type: string
example: john@smith.com
message:
type: string
nullable: true
example: >-
We successfully processed your meter reading of 21 kWh dated Dec 31,
2022
template:
type: string
example: Dear John,
%MESSAGE%
Mike from your service team
nullable: true
sources:
type: array
description: Any sources that were used to create the result
items:
$ref: '#/components/schemas/Source'
txId:
type: string
example: c916167c94
description: Unique transaction id. Corresponds to a log entry in enneo
Source:
type: object
description: A source that was used by an AI model / intent
properties:
type:
type: string
description: >
The type of source used. ticket = A similar ticket existed and that
response was used as a basis
knowledgeSource = An external knowledge source was used, usually a
company Wiki or FAQ
template = A predefined text template could be applied
languageModel = If the AI model did not use anything end created the
answer by itself
enum:
- ticket
- knowledgeSource
- template
- languageModel
example: knowledgeSource
id:
type: integer
description: Reference to the entry id, e.g. ticketId (if type=ticket)
example: 376189
nullable: true
name:
type: string
description: Name of reference (e.g. FAQ title or ticket title)
example: Opening hours
url:
type: string
description: >-
URL to associated source. null for sources with internal ids
(tickets and templates) or if non-existant (language model source)
example: https://company.com/faq/376189
nullable: true
text:
type: string
description: Full text that was used as source
example: >-
Our service hours are from 8am to 5pm. We are closed on weekends.
[...]
securitySchemes:
bearerAuth:
type: http
scheme: bearer
bearerFormat: JWT
description: JWT-based authentication
x-scopes:
api: Full access to the API
cookieAuth:
type: apiKey
in: cookie
name: connect.sid
description: Cookie-based authentication
x-scopes:
api: Full access to the API
````