naftiko: 1.0.0-alpha2 info: label: Deepgram Speech-to-Text API — Pre-Recorded description: 'Deepgram Speech-to-Text API — Pre-Recorded. 1 operations. Lead operation: Deepgram Transcribe pre-recorded audio. Self-contained Naftiko capability covering one Deepgram business surface.' tags: - Deepgram - Pre-Recorded created: '2026-05-19' modified: '2026-05-19' binds: - namespace: env keys: DEEPGRAM_API_KEY: DEEPGRAM_API_KEY capability: consumes: - type: http namespace: speech-to-text-pre-recorded baseUri: https://api.deepgram.com description: Deepgram Speech-to-Text API — Pre-Recorded business capability. Self-contained, no shared references. resources: - name: v1-listen path: /v1/listen operations: - name: transcribeprerecordedaudio method: POST description: Deepgram Transcribe pre-recorded audio outputRawFormat: json outputParameters: - name: result type: object value: $. inputParameters: - name: body in: body type: object description: Request body (JSON). required: true authentication: type: bearer token: '{{env.DEEPGRAM_API_KEY}}' exposes: - type: rest namespace: speech-to-text-pre-recorded-rest port: 8080 description: REST adapter for Deepgram Speech-to-Text API — Pre-Recorded. One Spectral-compliant resource per consumed operation, prefixed with /v1. resources: - path: /v1/v1/listen name: v1-listen description: REST surface for v1-listen. operations: - method: POST name: transcribeprerecordedaudio description: Deepgram Transcribe pre-recorded audio call: speech-to-text-pre-recorded.transcribeprerecordedaudio with: body: rest.body outputParameters: - type: object mapping: $. - type: mcp namespace: speech-to-text-pre-recorded-mcp port: 9090 transport: http description: MCP adapter for Deepgram Speech-to-Text API — Pre-Recorded. One tool per consumed operation, routed inline through this capability's consumes block. tools: - name: deepgram-transcribe-pre-recorded-audio description: Deepgram Transcribe pre-recorded audio hints: readOnly: false destructive: false idempotent: false call: speech-to-text-pre-recorded.transcribeprerecordedaudio with: body: tools.body outputParameters: - type: object mapping: $.