Prezent Streaming API

openapi: 3.1.0
info:
  title: Prezent Platform Audiences Streaming API
  version: 1.0.0
  summary: Agent-ready REST API for Prezent's content-generation and template-conversion services.
  description: "The Prezent Platform API exposes Prezent's AutoGenerator, Template Converter,\nAudiences, Themes, and File-upload services as a uniform JSON HTTP API.\n\nThis contract describes the **target agent-ready shape** of the API: every\ndocumented endpoint returns a uniform success envelope `{ \"success\": true,\n\"data\": { ... } }` or a uniform error envelope `{ \"success\": false,\n\"data\": null, \"error\": { \"code\": \"...\", \"message\": \"...\", \"details\": { ... } } }`.\nError codes are drawn from a stable catalog (see `ErrorCode`). The contract\nis additive: handlers may continue to emit legacy keys alongside the\ndocumented ones for backwards compatibility, so schemas explicitly allow\nadditional properties at the envelope and data levels.\n\nSeveral endpoints kick off long-running background work and return a\n`callback_id` that the client polls against a corresponding status\nendpoint. The canonical async polling endpoint for the Template Converter\nfamily is **`GET /api/v2/template-converter/status/{callback_id}`** which\nuses strict HTTP-status mapping (200 only on workflow success, 4xx/5xx on\nworkflow failure). The v1 polling endpoint at\n`/api/v1/template-converter/status/{callbackId}` (and its REST alias\n`GET /api/v1/template-conversions/{callback_id}`) continues to be served\nwith legacy semantics (HTTP 200 with `status` field) and is intentionally\nomitted from this documentation.\n\n**REST resource aliases (v1.1).** AutoGenerator and Template Converter\nendpoints are documented here under their REST-resource path names\n(`/api/v1/autogenerations`, `/api/v1/template-conversions`). The legacy\nRPC-style paths under `/api/v1/autogenerator/*` and\n`/api/v1/template-converter/*` continue to be served indefinitely; they\nreturn a `Deprecation: true` header pointing to the successor alias.\n\n**Conventions that apply to every endpoint** (full reference in the\ncompanion guides on this documentation site):\n\n- **Auth:** API key as a Bearer token — `Authorization: Bearer <key>`.\n  Each key is scoped to an allow-list of endpoint paths; an out-of-scope\n  call returns `404 ENDPOINT_NOT_FOUND`.\n- **Rate limits:** gateway throttle 10 req/s sustained, 5 burst,\n  1,000 req/day per key; per-company per-category 60-second sliding\n  windows; annual quotas (50,000 slide generations/yr, 1,000,000\n  downloads/yr). Limit breaches return `429` with `error.code`\n  `TOO_MANY_REQUESTS` (gateway), `RATE_LIMIT_EXCEEDED` (per-category),\n  or `USAGE_LIMIT_EXCEEDED` (annual).\n- **Errors:** every error uses the `{ success:false, data:null, error:{\n  code, message, details } }` envelope with stable, documented `code`\n  values.\n- **Idempotency & pagination:** mutating requests accept an\n  `Idempotency-Key`; list endpoints accept opt-in `limit`/`cursor`\n  paging.\n\n**Building with an AI agent?** Prezent ships a first-class Model Context\nProtocol (MCP) server. The hosted endpoint is\n`https://mcp.myprezent.com/mcp` (OAuth 2.1, works with Claude.ai Custom\nConnectors); a local stdio server is published on PyPI as\n`prezent-mcp-server` and plugs into Claude Desktop, Cursor, Cline,\nContinue, and Zed. The MCP tools wrap the same REST endpoints\ndescribed here.\n\n**Official SDKs.** Typed REST clients are published for Python and\nTypeScript (`prezent-sdk`) — generated from and versioned against this\nspec. For other languages, generate a client from this document.\n\n**Out of scope of this document:** SCIM endpoints under `/api/v1/scim/*`.\nThose endpoints conform to RFC 7644 (SCIM 2.0) and follow a different\nenvelope (`schemas`, `Resources`, `totalResults`, etc.). They are\ndocumented separately at /docs/scim-user-management.\n"
  x-out-of-scope-note: 'SCIM endpoints (`/api/v1/scim/*`) are NOT included in this spec. They

    follow RFC 7644 (SCIM 2.0) and are documented in a separate document at

    /docs/scim-user-management.

    '
  contact:
    name: Prezent API Support
    email: support@prezent.ai
    url: https://prezent.ai
  license:
    name: Proprietary - Prezent, Inc.
    url: https://prezent.ai/terms
servers:
- url: https://api.prezent.ai
  description: Production (canonical)
- url: https://uatstage-api.myprezent.com
  description: UAT (user acceptance testing)
- url: https://devstage-api.myprezent.com
  description: Development
security:
- BearerAuth: []
tags:
- name: Streaming
  description: Open a Server-Sent Events stream to receive real-time progress events as Prezent generates a presentation. See the [Streaming guide](/docs/streaming) for a full walk-through plus reconnect semantics.
paths:
  /api/v1/streams/sessions:
    post:
      tags:
      - Streaming
      summary: Open a stream session for an existing job
      description: 'Returns a short-lived URL (5-minute TTL) that you open with

        EventSource (or any SSE client) to receive a live feed of

        progress events for the job identified by `callback_id`. The

        token in the URL is bound to your API key and to that

        `callback_id` only.


        See the [Streaming guide](/docs/streaming) for the full

        walk-through, the event catalog, and reconnect semantics.

        '
      operationId: createStreamSession
      x-rate-limit-category: general
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/StreamSessionCreateRequest'
      responses:
        '201':
          description: A streaming URL has been minted.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/StreamSessionCreateResponse'
        '400':
          $ref: '#/components/responses/BadRequest'
        '401':
          $ref: '#/components/responses/Unauthorized'
        '403':
          $ref: '#/components/responses/Forbidden'
        '404':
          $ref: '#/components/responses/NotFound'
        '429':
          $ref: '#/components/responses/RateLimited'
        '500':
          $ref: '#/components/responses/InternalServerError'
        '503':
          $ref: '#/components/responses/ServiceUnavailable'
        '504':
          $ref: '#/components/responses/GatewayTimeout'
components:
  responses:
    RateLimited:
      description: "Rate limit, usage limit, or gateway-level throttle exceeded.\n`error.code` is `TOO_MANY_REQUESTS` (gateway throttle),\n`RATE_LIMIT_EXCEEDED` (per-category), or `USAGE_LIMIT_EXCEEDED`\n(annual quota).\n\nDefault limits (all configurable per company/key):\n- Gateway throttle (per API key) → `TOO_MANY_REQUESTS`: 10 requests/second\n  sustained, 5 burst, 1,000 requests/day.\n- Per-company, per-category sliding 60-second window →\n  `RATE_LIMIT_EXCEEDED`. The applicable category is given by each\n  operation's `x-rate-limit-category`.\n- Annual usage quota → `USAGE_LIMIT_EXCEEDED`: 50,000 slide\n  generations/year and 1,000,000 presentation downloads/year.\n\n`X-RateLimit-Limit`/`X-RateLimit-Remaining`/`X-RateLimit-Reset` are\nreturned on successful (2xx) responses from rate-limited endpoints\nand, with `Retry-After`, on the per-category `RATE_LIMIT_EXCEEDED`\n429 (the headers declared below). The gateway `TOO_MANY_REQUESTS`\nand annual `USAGE_LIMIT_EXCEEDED` responses do not carry them. Read\n`X-RateLimit-Remaining` to self-throttle and honour `Retry-After` on\na 429.\n"
      headers:
        X-RateLimit-Limit:
          $ref: '#/components/headers/XRateLimitLimit'
        X-RateLimit-Remaining:
          $ref: '#/components/headers/XRateLimitRemaining'
        X-RateLimit-Reset:
          $ref: '#/components/headers/XRateLimitReset'
        Retry-After:
          $ref: '#/components/headers/RetryAfter'
      content:
        application/json:
          schema:
            $ref: '#/components/schemas/ErrorEnvelope'
          example:
            success: false
            data: null
            error:
              code: RATE_LIMIT_EXCEEDED
              message: Rate limit exceeded for this category. Try again later.
              details:
                category: auto_generator
                retry_after_seconds: 12
    ServiceUnavailable:
      description: Service is temporarily unavailable (downstream dependency unhealthy). `error.code` is `SERVICE_UNAVAILABLE` or `EXTERNAL_SERVICE_ERROR`.
      content:
        application/json:
          schema:
            $ref: '#/components/schemas/ErrorEnvelope'
          example:
            success: false
            data: null
            error:
              code: SERVICE_UNAVAILABLE
              message: Service is temporarily unavailable.
              details: {}
    InternalServerError:
      description: Unexpected server error. `error.code` is `INTERNAL_SERVER_ERROR`.
      content:
        application/json:
          schema:
            $ref: '#/components/schemas/ErrorEnvelope'
          example:
            success: false
            data: null
            error:
              code: INTERNAL_SERVER_ERROR
              message: An unexpected server error occurred.
              details: {}
    Unauthorized:
      description: 'Caller did not present a valid Bearer token, or the token has

        expired. `error.code` is one of `UNAUTHORIZED`, `INVALID_API_KEY`,

        `EXPIRED_API_KEY`.

        '
      content:
        application/json:
          schema:
            $ref: '#/components/schemas/ErrorEnvelope'
          example:
            success: false
            data: null
            error:
              code: INVALID_API_KEY
              message: API key not found or not authorized.
              details: {}
    BadRequest:
      description: Generic client error. `error.code` is one of `BAD_REQUEST`, `INVALID_JSON`, `MISSING_REQUIRED_FIELD`, `MISSING_QUERY_PARAM`, `MISSING_CALLBACK_ID`, `MISSING_SLIDES_ARRAY`, `MISSING_PROMPT`, `MISSING_TEMPLATE_ID`, `MISSING_FILE_CONTENT`, `MISSING_SHARE_DETAILS`, `INVALID_TYPE`, `INVALID_DATA`, `INVALID_DATA_TYPE`, `INVALID_PAYLOAD`, `INVALID_REQUEST`, `API_REQUEST_FAILED`, or `FILE_UPLOAD_FAILED`.
      content:
        application/json:
          schema:
            $ref: '#/components/schemas/ErrorEnvelope'
          example:
            success: false
            data: null
            error:
              code: BAD_REQUEST
              message: Bad request.
              details: {}
    NotFound:
      description: 'Requested endpoint or resource does not exist. `error.code` is

        one of `ENDPOINT_NOT_FOUND`, `RESOURCE_NOT_FOUND`, `NOT_FOUND`.

        '
      content:
        application/json:
          schema:
            $ref: '#/components/schemas/ErrorEnvelope'
          example:
            success: false
            data: null
            error:
              code: RESOURCE_NOT_FOUND
              message: The requested resource was not found.
              details: {}
    Forbidden:
      description: Caller is authenticated but not allowed to perform this operation. `error.code` is `FORBIDDEN`.
      content:
        application/json:
          schema:
            $ref: '#/components/schemas/ErrorEnvelope'
          example:
            success: false
            data: null
            error:
              code: FORBIDDEN
              message: Access to the resource is forbidden.
              details: {}
    GatewayTimeout:
      description: A downstream call timed out. `error.code` is `GATEWAY_TIMEOUT`.
      content:
        application/json:
          schema:
            $ref: '#/components/schemas/ErrorEnvelope'
          example:
            success: false
            data: null
            error:
              code: GATEWAY_TIMEOUT
              message: Gateway timeout.
              details: {}
  headers:
    RetryAfter:
      description: Number of seconds the client should wait before retrying.
      schema:
        type: integer
        format: int32
        minimum: 0
    XRateLimitRemaining:
      description: 'Approximate number of requests remaining in the current

        60-second window for this caller and category.

        '
      schema:
        type: integer
        format: int32
        minimum: 0
    XRateLimitReset:
      description: Unix timestamp (seconds) when the current rate-limit window resets.
      schema:
        type: integer
        format: int64
        minimum: 0
    XRateLimitLimit:
      description: 'The configured request limit for the rate-limit category

        (requests per 60-second window) that applied to this call.

        The applicable category is the operation''s `x-rate-limit-category`.

        Returned on every response (success and 429) so clients can

        self-throttle. See the Rate limits guide for the default

        per-category numbers.

        '
      schema:
        type: integer
        format: int32
        minimum: 0
  schemas:
    StreamSessionCreateRequest:
      type: object
      required:
      - callback_id
      properties:
        callback_id:
          type: string
          description: 'The `callback_id` returned by your earlier call to

            `POST /api/v1/autogenerations` (or other async endpoint).

            The calling API key must own this job.

            '
          example: cb_01HZQ3Y7K3D6V0A0CS7HD3
    StreamSessionCreateResponse:
      type: object
      required:
      - success
      - data
      properties:
        success:
          type: boolean
          example: true
        data:
          type: object
          required:
          - session_id
          - stream_url
          - expires_at
          properties:
            session_id:
              type: string
              description: Same as the `callback_id` you passed in.
              example: cb_01HZQ3Y7K3D6V0A0CS7HD3
            stream_url:
              type: string
              format: uri
              description: 'Open with `new EventSource(stream_url)` (or `fetch`

                + ReadableStream). Hostname is environment-dependent

                (e.g. `stream-api.myprezent.com`). The embedded token

                expires at `expires_at`.

                '
              example: https://stream-api.myprezent.com/v1/streams/cb_01HZQ3Y7K3D6V0A0CS7HD3?token=eyJhbGciOi...
            expires_at:
              type: string
              format: date-time
              description: When the embedded token stops being accepted. Call this endpoint again to refresh.
              example: '2026-06-09T10:05:00Z'
    ErrorEnvelope:
      type: object
      description: 'Standard error envelope returned by every documented endpoint

        (both in-Lambda and gateway-level errors). Additional legacy

        keys may appear alongside `success`/`data`/`error` for

        backwards compatibility.

        '
      required:
      - success
      - data
      - error
      properties:
        success:
          type: boolean
          const: false
          description: Always `false` on error.
        data:
          type:
          - object
          - 'null'
          description: Always `null` on error.
        error:
          type: object
          required:
          - code
          - message
          description: Error description.
          properties:
            code:
              $ref: '#/components/schemas/ErrorCode'
            message:
              type: string
              description: Human-readable error description (English).
            details:
              type: object
              description: Optional structured details (validation failures, downstream-service info, etc.).
              additionalProperties: true
      additionalProperties: true
    ErrorCode:
      type: string
      description: 'Stable identifier for an error condition. Codes never change

        meaning or HTTP status once published.

        '
      enum:
      - INVALID_API_KEY
      - EXPIRED_API_KEY
      - UNAUTHORIZED
      - FORBIDDEN
      - ENDPOINT_NOT_FOUND
      - RESOURCE_NOT_FOUND
      - NOT_FOUND
      - METHOD_NOT_ALLOWED
      - UNSUPPORTED_FILE_TYPE
      - INVALID_INPUT
      - UNPROCESSABLE_ENTITY
      - RATE_LIMIT_EXCEEDED
      - USAGE_LIMIT_EXCEEDED
      - TOO_MANY_REQUESTS
      - THROTTLED
      - BAD_REQUEST
      - INVALID_JSON
      - MISSING_REQUIRED_FIELD
      - MISSING_QUERY_PARAM
      - MISSING_CALLBACK_ID
      - MISSING_SLIDES_ARRAY
      - MISSING_PROMPT
      - MISSING_TEMPLATE_ID
      - MISSING_FILE_CONTENT
      - MISSING_SHARE_DETAILS
      - INVALID_TYPE
      - INVALID_DATA
      - INVALID_DATA_TYPE
      - INVALID_PAYLOAD
      - INVALID_REQUEST
      - API_REQUEST_FAILED
      - FILE_UPLOAD_FAILED
      - MAX_RETRIES_EXCEEDED
      - EXTERNAL_SERVICE_ERROR
      - INTERNAL_SERVER_ERROR
      - SERVICE_UNAVAILABLE
      - GATEWAY_TIMEOUT
  securitySchemes:
    BearerAuth:
      type: http
      scheme: bearer
      bearerFormat: opaque
      description: 'Bearer token authentication. Supply the API key issued by

        Customer Success as `Authorization: Bearer <api_key>`. The API

        does NOT accept the AWS `x-api-key` header. Keys may have an

        expiry date; expired keys return `401 EXPIRED_API_KEY`. Keys are

        scoped to specific paths; calling an unauthorised path returns

        `404 ENDPOINT_NOT_FOUND`.

        '