naftiko: 1.0.0-alpha2
info:
  label: Hugging Face Inference API — Computer Vision
  description: 'Hugging Face Inference API — Computer Vision. 2 operations. Lead operation: Image Classification Inference.
    Self-contained Naftiko capability covering one Hugging Face business surface.'
  tags:
  - Hugging Face
  - Computer Vision
  created: '2026-05-19'
  modified: '2026-05-19'
binds:
- namespace: env
  keys:
    HUGGING_FACE_API_KEY: HUGGING_FACE_API_KEY
capability:
  consumes:
  - type: http
    namespace: inference-computer-vision
    baseUri: https://api-inference.huggingface.co
    description: Hugging Face Inference API — Computer Vision business capability. Self-contained, no shared references.
    resources:
    - name: models-model_id-image-classification
      path: /models/{model_id}/image-classification
      operations:
      - name: imageclassification
        method: POST
        description: Image Classification Inference
        outputRawFormat: json
        outputParameters:
        - name: result
          type: object
          value: $.
        inputParameters:
        - name: model_id
          in: path
          type: string
          required: true
        - name: body
          in: body
          type: object
          description: Request body (JSON).
          required: true
    - name: models-model_id-object-detection
      path: /models/{model_id}/object-detection
      operations:
      - name: objectdetection
        method: POST
        description: Object Detection Inference
        outputRawFormat: json
        outputParameters:
        - name: result
          type: object
          value: $.
        inputParameters:
        - name: model_id
          in: path
          type: string
          required: true
        - name: body
          in: body
          type: object
          description: Request body (JSON).
          required: true
    authentication:
      type: bearer
      token: '{{env.HUGGING_FACE_API_KEY}}'
  exposes:
  - type: rest
    namespace: inference-computer-vision-rest
    port: 8080
    description: REST adapter for Hugging Face Inference API — Computer Vision. One Spectral-compliant resource per consumed
      operation, prefixed with /v1.
    resources:
    - path: /v1/models/{model-id}/image-classification
      name: models-model-id-image-classification
      description: REST surface for models-model_id-image-classification.
      operations:
      - method: POST
        name: imageclassification
        description: Image Classification Inference
        call: inference-computer-vision.imageclassification
        with:
          model_id: rest.model_id
          body: rest.body
        outputParameters:
        - type: object
          mapping: $.
    - path: /v1/models/{model-id}/object-detection
      name: models-model-id-object-detection
      description: REST surface for models-model_id-object-detection.
      operations:
      - method: POST
        name: objectdetection
        description: Object Detection Inference
        call: inference-computer-vision.objectdetection
        with:
          model_id: rest.model_id
          body: rest.body
        outputParameters:
        - type: object
          mapping: $.
  - type: mcp
    namespace: inference-computer-vision-mcp
    port: 9090
    transport: http
    description: MCP adapter for Hugging Face Inference API — Computer Vision. One tool per consumed operation, routed inline
      through this capability's consumes block.
    tools:
    - name: image-classification-inference
      description: Image Classification Inference
      hints:
        readOnly: false
        destructive: false
        idempotent: false
      call: inference-computer-vision.imageclassification
      with:
        model_id: tools.model_id
        body: tools.body
      outputParameters:
      - type: object
        mapping: $.
    - name: object-detection-inference
      description: Object Detection Inference
      hints:
        readOnly: false
        destructive: false
        idempotent: false
      call: inference-computer-vision.objectdetection
      with:
        model_id: tools.model_id
        body: tools.body
      outputParameters:
      - type: object
        mapping: $.