naftiko: 1.0.0-alpha2 info: label: Microsoft Windows 10 Windows ML API — Models description: 'Microsoft Windows 10 Windows ML API — Models. 4 operations. Lead operation: Microsoft Windows 10 List loaded ML models. Self-contained Naftiko capability covering one Microsoft Windows 10 business surface.' tags: - Microsoft Windows 10 - Models created: '2026-05-19' modified: '2026-05-19' binds: - namespace: env keys: MICROSOFT_WINDOWS_10_API_KEY: MICROSOFT_WINDOWS_10_API_KEY capability: consumes: - type: http namespace: ml-models baseUri: https://api.windows.com description: Microsoft Windows 10 Windows ML API — Models business capability. Self-contained, no shared references. resources: - name: ml-models path: /ml/models operations: - name: listmlmodels method: GET description: Microsoft Windows 10 List loaded ML models outputRawFormat: json outputParameters: - name: result type: object value: $. - name: loadmlmodel method: POST description: Microsoft Windows 10 Load a machine learning model outputRawFormat: json outputParameters: - name: result type: object value: $. inputParameters: - name: body in: body type: object description: Request body (JSON). required: true - name: ml-models-modelId path: /ml/models/{modelId} operations: - name: getmlmodel method: GET description: Microsoft Windows 10 Get model details outputRawFormat: json outputParameters: - name: result type: object value: $. inputParameters: - name: modelId in: path type: string description: Unique identifier for the loaded model required: true - name: unloadmlmodel method: DELETE description: Microsoft Windows 10 Unload a machine learning model outputRawFormat: json outputParameters: - name: result type: object value: $. inputParameters: - name: modelId in: path type: string description: Unique identifier for the model required: true exposes: - type: rest namespace: ml-models-rest port: 8080 description: REST adapter for Microsoft Windows 10 Windows ML API — Models. One Spectral-compliant resource per consumed operation, prefixed with /v1. resources: - path: /v1/ml/models name: ml-models description: REST surface for ml-models. operations: - method: GET name: listmlmodels description: Microsoft Windows 10 List loaded ML models call: ml-models.listmlmodels outputParameters: - type: object mapping: $. - method: POST name: loadmlmodel description: Microsoft Windows 10 Load a machine learning model call: ml-models.loadmlmodel with: body: rest.body outputParameters: - type: object mapping: $. - path: /v1/ml/models/{modelid} name: ml-models-modelid description: REST surface for ml-models-modelId. operations: - method: GET name: getmlmodel description: Microsoft Windows 10 Get model details call: ml-models.getmlmodel with: modelId: rest.modelId outputParameters: - type: object mapping: $. - method: DELETE name: unloadmlmodel description: Microsoft Windows 10 Unload a machine learning model call: ml-models.unloadmlmodel with: modelId: rest.modelId outputParameters: - type: object mapping: $. - type: mcp namespace: ml-models-mcp port: 9090 transport: http description: MCP adapter for Microsoft Windows 10 Windows ML API — Models. One tool per consumed operation, routed inline through this capability's consumes block. tools: - name: microsoft-windows-10-list-loaded description: Microsoft Windows 10 List loaded ML models hints: readOnly: true destructive: false idempotent: true call: ml-models.listmlmodels outputParameters: - type: object mapping: $. - name: microsoft-windows-10-load-machine description: Microsoft Windows 10 Load a machine learning model hints: readOnly: false destructive: false idempotent: false call: ml-models.loadmlmodel with: body: tools.body outputParameters: - type: object mapping: $. - name: microsoft-windows-10-get-model description: Microsoft Windows 10 Get model details hints: readOnly: true destructive: false idempotent: true call: ml-models.getmlmodel with: modelId: tools.modelId outputParameters: - type: object mapping: $. - name: microsoft-windows-10-unload-machine description: Microsoft Windows 10 Unload a machine learning model hints: readOnly: false destructive: true idempotent: true call: ml-models.unloadmlmodel with: modelId: tools.modelId outputParameters: - type: object mapping: $.