# Generated by API Evangelist (build-phrasing.py). Our phrasing, not observed demand.
overlay: 1.0.0
info:
  title: API Evangelist conversational phrasing for Microsoft Azure Predictions API
  version: 1.0.0
extends: openapi/microsoft-azure-predictionsapi-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: 4
- target: $.paths['/projects/{projectId}/predictions'].delete
  update:
    x-apievangelist-phrasing:
      intent: Delete stored prediction images
      effect: destructive
      questions:
      - How do I delete predicted images and their results from a Custom Vision project?
      - Can I clear out specific stored predictions by id?
      instructions:
      - text: Delete predictions {ids} from project {projectId}.
        slots:
          projectId: path.projectId
          ids: query.ids
      - text: Remove the stored prediction images {ids} and their results in project {projectId}.
        slots:
          projectId: path.projectId
          ids: query.ids
      method: generated
      generated: '2026-09-26'
- target: $.paths['/projects/{projectId}/predictions/query'].post
  update:
    x-apievangelist-phrasing:
      intent: Query images sent to the prediction endpoint
      effect: read
      questions:
      - Which images were sent to my Custom Vision prediction endpoint?
      - Can I filter stored predictions by iteration, tag or time range?
      instructions:
      - text: Show images sent to the prediction endpoint for project {projectId}.
        slots:
          projectId: path.projectId
      - text: Query predictions in project {projectId} for iteration {iterationId} between {startTime} and {endTime}.
        slots:
          projectId: path.projectId
          iterationId: requestBody.iterationId
          startTime: requestBody.startTime
          endTime: requestBody.endTime
      - text: Get up to {maxCount} stored predictions tagged {tags} in project {projectId}.
        slots:
          projectId: path.projectId
          maxCount: requestBody.maxCount
          tags: requestBody.tags
      method: generated
      generated: '2026-09-26'
- target: $.paths['/projects/{projectId}/quicktest/image'].post
  update:
    x-apievangelist-phrasing:
      intent: Quick test an uploaded image
      effect: write
      questions:
      - How do I quickly test my trained model against an image file I upload?
      - Can I quick test an image without storing the result?
      instructions:
      - text: Upload image file {imageData} and quick test it against project {projectId}.
        slots:
          projectId: path.projectId
          imageData: requestBody.imageData
      - text: Quick test image file {imageData} on iteration {iterationId} of project {projectId}.
        slots:
          projectId: path.projectId
          imageData: requestBody.imageData
          iterationId: query.iterationId
      method: generated
      generated: '2026-09-26'
- target: $.paths['/projects/{projectId}/quicktest/url'].post
  update:
    x-apievangelist-phrasing:
      intent: Quick test an image by URL
      effect: write
      questions:
      - Can I test my Custom Vision model with an image hosted at a URL?
      - How do I try a specific iteration against a remote image link?
      instructions:
      - text: Quick test the image at {url} against project {projectId}.
        slots:
          projectId: path.projectId
          url: requestBody.url
      - text: Quick test image URL {url} on iteration {iterationId} of project {projectId}.
        slots:
          projectId: path.projectId
          url: requestBody.url
          iterationId: query.iterationId
      method: generated
      generated: '2026-09-26'