> ## Documentation Index > Fetch the complete documentation index at: https://developer.demand-iq.com/llms.txt > Use this file to discover all available pages before exploring further. # Answer a question > Answers a viewer question. Checks FAQs first (exact then semantic match at >= 0.6 confidence), falls back to GPT-4o with slide context. Returns answer text plus synthesized audio. `deckId` and `presentationId` must be valid UUIDs; other values are rejected with a 400. When a `presentationId` is provided, the answer is drawn from the deck that presentation belongs to; the body's `deckId` is used only as a fallback when that deck has been deleted. If `slideId` doesn't match a slide in the deck, the question is answered without slide context instead of failing. **Authentication**: Accepts either a session cookie (admin) or a `presentationId` (public viewer). ## OpenAPI ````yaml /openapi.json post /api/qa openapi: 3.0.0 info: title: AI Presentations API version: 1.0.0 description: >- REST API for AI Presentations — an AI-powered presentation platform with narrated slides, live Q&A, and deck management. **Authentication**: Most endpoints require a session cookie obtained from `POST /api/auth/login`. Presentation playback endpoints (`/api/qa`, `/api/narration/*`) also accept a `presentationId` for unauthenticated viewer access. contact: name: Demand IQ url: https://demand-iq.com servers: - url: https://app.demand-iq.com description: Production - url: http://localhost:3000 description: Local development security: [] tags: - name: Decks description: Manage presentation decks - name: Slides description: Manage slides within a deck - name: FAQs description: Manage pre-built Q&A pairs - name: Branding description: Colors, logo, and call-to-action configuration - name: Q&A description: Live question answering and Q&A settings - name: Narration description: Text-to-speech synthesis - name: Presentations description: Create and manage personalized presentation instances - name: Generation description: AI-powered content generation from slide images - name: Images description: Image upload and management - name: Voices description: Available TTS voices - name: Utility description: Health check and diagnostics - name: Authentication description: Session-based login, logout, and identity - name: Company description: Organization contact info and knowledge base - name: Notifications description: Event subscriptions and delivery log - name: Contracts description: Contract template import and e-signature - name: Fonts description: Available Google Fonts for branding - name: Roof Measurements description: Request and track automated roof measurements - name: Actions description: Voice/text command detection for presentation control - name: Appointments description: Appointment scheduling for presentation viewers - name: Activity description: Recent presentation activity feed - name: Onboarding description: >- Per-user product onboarding milestones (ENG-641). Milestone endpoints are idempotent and are called by the app UI — they should not be treated as tenant-integration surface. paths: /api/qa: post: tags: - Q&A summary: Answer a question description: >- Answers a viewer question. Checks FAQs first (exact then semantic match at >= 0.6 confidence), falls back to GPT-4o with slide context. Returns answer text plus synthesized audio. `deckId` and `presentationId` must be valid UUIDs; other values are rejected with a 400. When a `presentationId` is provided, the answer is drawn from the deck that presentation belongs to; the body's `deckId` is used only as a fallback when that deck has been deleted. If `slideId` doesn't match a slide in the deck, the question is answered without slide context instead of failing. **Authentication**: Accepts either a session cookie (admin) or a `presentationId` (public viewer). requestBody: required: true content: application/json: schema: type: object required: - question - slideId properties: question: type: string minLength: 3 slideId: type: string minLength: 1 maxLength: 100 description: >- Current slide ID for context. Accepts a slide UUID or a legacy semantic ID such as `slide-1`. deckId: type: string format: uuid description: >- Deck to answer from. Ignored when `presentationId` is provided — the deck is resolved from the presentation, and this value is used only as a fallback if that deck has been deleted. presentationId: type: string format: uuid description: Required for unauthenticated viewer access history: type: array items: type: object required: - role - content properties: role: type: string enum: - user - assistant content: type: string voiceId: type: string language: type: string example: en description: BCP-47 language tag responses: '200': description: Answer with audio content: application/json: schema: $ref: '#/components/schemas/QaResponse' '400': description: Bad request content: application/json: schema: $ref: '#/components/schemas/Error' '401': description: Unauthorized content: application/json: schema: $ref: '#/components/schemas/Error' '403': description: Forbidden content: application/json: schema: $ref: '#/components/schemas/Error' '404': description: Not found content: application/json: schema: $ref: '#/components/schemas/Error' '500': description: Internal server error content: application/json: schema: $ref: '#/components/schemas/Error' components: schemas: QaResponse: type: object properties: answer: type: string audioUrl: type: string description: Accessible URL for the TTS audio of the answer wordTimings: type: array items: $ref: '#/components/schemas/WordTiming' source: type: string enum: - faq - cached - openai - fallback relatedSlideId: type: string nullable: true required: - answer - audioUrl - source Error: type: object properties: error: type: string required: - error WordTiming: type: object properties: word: type: string startTime: type: number description: Start time in milliseconds endTime: type: number description: End time in milliseconds required: - word - startTime - endTime ````