# Generated by API Evangelist (build-phrasing.py). Our phrasing, not observed demand.
overlay: 1.0.0
info:
  title: API Evangelist conversational phrasing for Sensei Machine Learning ML Services API
  version: 1.0.0
extends: openapi/adobe-suite-mlservices-api-openapi.yml
actions:
- target: $.info
  update:
    x-apievangelist-phrasing:
      method: generated
      generated: '2026-09-26'
      generator: build-phrasing.py
      label: Generated by API Evangelist
      operations: 6
- target: $.paths['/mlServices'].get
  update:
    x-apievangelist-phrasing:
      intent: List MLServices
      effect: read
      questions:
      - Which machine learning services are deployed in my sandbox?
      - Can I search MLServices by name or tag?
      instructions:
      - text: List the MLServices in sandbox {sandbox}.
        slots:
          sandbox: header.x-sandbox-name
      - text: Find MLServices matching {property}, {limit} at a time.
        slots:
          property: query.property
          limit: query.limit
      method: generated
      generated: '2026-09-26'
- target: $.paths['/mlServices'].post
  update:
    x-apievangelist-phrasing:
      intent: Create an MLService
      effect: write
      questions:
      - How do I publish a trained ML instance as a service that scores on a schedule?
      - What is required to create an MLService, like the scoring dataset and schedule?
      instructions:
      - text: Create MLService {name} for ML instance {mlInstanceId} scoring dataset {scoringDataSetId} on schedule {scoringSchedule}.
        slots:
          name: requestBody.name
          mlInstanceId: requestBody.mlInstanceId
          scoringDataSetId: requestBody.scoringDataSetId
          scoringSchedule: requestBody.scoringSchedule
      - text: Set up service {name} on instance {mlInstanceId}, training on dataset {trainingDataSetId} with schedule {trainingSchedule}.
        slots:
          name: requestBody.name
          mlInstanceId: requestBody.mlInstanceId
          trainingDataSetId: requestBody.trainingDataSetId
          trainingSchedule: requestBody.trainingSchedule
      method: generated
      generated: '2026-09-26'
- target: $.paths['/mlServices'].delete
  update:
    x-apievangelist-phrasing:
      intent: Delete all MLServices for an ML instance
      effect: destructive
      questions:
      - Can I remove every MLService built from one ML instance at once?
      - What wipes out all services attached to an instance?
      instructions:
      - text: Delete all MLServices for ML instance {mlInstanceId}.
        slots:
          mlInstanceId: query.mlInstanceId
      - text: Remove every service created from instance {mlInstanceId}.
        slots:
          mlInstanceId: query.mlInstanceId
      method: generated
      generated: '2026-09-26'
- target: $.paths['/mlServices/{serviceId}'].get
  update:
    x-apievangelist-phrasing:
      intent: Get one MLService
      effect: read
      questions:
      - How do I see the schedules and datasets of a single MLService?
      - What does one ML service look like when fetched by ID?
      instructions:
      - text: Show MLService {serviceId}.
        slots:
          serviceId: path.serviceId
      - text: Get the training and scoring setup of service {serviceId}.
        slots:
          serviceId: path.serviceId
      method: generated
      generated: '2026-09-26'
- target: $.paths['/mlServices/{serviceId}'].put
  update:
    x-apievangelist-phrasing:
      intent: Update an MLService
      effect: write
      questions:
      - Can I change the scoring schedule of an existing MLService?
      - Is there optimistic locking with an ETag when updating a service?
      instructions:
      - text: Update MLService {serviceId} to score on schedule {scoringSchedule}.
        slots:
          serviceId: path.serviceId
          scoringSchedule: requestBody.scoringSchedule
      - text: Point existing service {serviceId} at scoring dataset {scoringDataSetId} for instance {mlInstanceId}.
        slots:
          serviceId: path.serviceId
          scoringDataSetId: requestBody.scoringDataSetId
          mlInstanceId: requestBody.mlInstanceId
      method: generated
      generated: '2026-09-26'
- target: $.paths['/mlServices/{serviceId}'].delete
  update:
    x-apievangelist-phrasing:
      intent: Delete one MLService
      effect: destructive
      questions:
      - How do I remove a single ML service I no longer need?
      - Does deleting one service leave the others on the same instance intact?
      instructions:
      - text: Delete MLService {serviceId}.
        slots:
          serviceId: path.serviceId
      - text: Remove just the service {serviceId}.
        slots:
          serviceId: path.serviceId
      method: generated
      generated: '2026-09-26'