openapi: 3.0.3
info:
  title: Transcripts API
  description: 'The Transcripts API enables you to send and retrieve transcript messages associated with interactions or workflow
    sessions in the Avaya Infinity contact center platform.


    ## Endpoints


    ### Send Transcript Batch

    `POST /api/transcripts/v1/transcripts/{id}`

    Submit a batch of transcript messages for a specific interaction or workflow session.

    [View API Reference →](https://developers.avayacloud.com/avaya-infinity/reference/transcripts-sendBatch)


    ### Get Transcripts

    `GET /api/transcripts/v1/transcripts/{interactionId}`

    Retrieve all transcript messages for a given interaction or workflow session.

    [View API Reference →](https://developers.avayacloud.com/avaya-infinity/reference/transcripts-get)


    ## Getting Started


    ### Find Your Customer Subdomain


    Your subdomain is found in your Avaya Infinity portal URL and is required for all API calls.


    **Example:** If your portal URL is:

    ```

    https://core.avaya1234.ec.avayacloud.com/app/core-config-ui/

    ```

    Your subdomain is: **`avaya1234`**


    ### Authentication


    To get started, contact Avaya Support to request a `client_id` and `client_secret` provisioned with the `TRANSCRIPT` scope.


    Once you have your credentials, [generate an access token](https://developers.avayacloud.com/avaya-infinity/reference/access-token)
    and include it in the `Authorization: Bearer <token>` header of each request.


    ### Working with Interactions and Sessions


    Both endpoints accept either an **interactionId** (for standard voice or digital interactions) or a **workflow session
    ID** (for `WORKFLOW_SESSION` type sessions). Use whichever ID corresponds to the context you are working with.


    ### Important Notes


    - The POST endpoint processes messages sequentially. Partial success is possible — inspect `failed` and `errors` in the
    response.

    - The `message` field has different types in POST vs GET: a plain string in the request, and an array of content objects
    in the response.

    - The `createdAt` field in GET responses uses the format `2026-02-26 16:27:05.124000` (space separator, no `T`, no timezone)
    rather than standard ISO 8601.

    - If the `{interactionId}` provided to the GET endpoint does not exist, the API currently returns a `500` error. This
    is a known issue pending a server-side fix.'
  contact:
    name: Avaya Developer Support
    url: https://developers.avayacloud.com/avaya-infinity/
  license:
    name: Avaya Software Development Kit (SDK) Software License Terms
    url: http://support.avaya.com/css/P8/documents/101038288
  version: 1.0.0
servers:
- url: https://core.{customerId}.ec.avayacloud.com/api/transcripts/v1
  description: Production
  variables:
    customerId:
      description: Your organization subdomain identifier (e.g., avaya1234)
      default: avaya1234
tags:
- name: Transcripts
  description: Endpoints for creating and retrieving transcript messages
paths:
  /transcripts/{id}:
    post:
      tags:
      - Transcripts
      summary: Send transcript batch
      operationId: transcripts.sendBatch
      security:
      - Bearer: []
      description: "Submits a batch of transcript messages for a given interaction or workflow session.\n\nThe `{id}` path\
        \ parameter accepts either an **interactionId** or a **workflow session ID**.\n\nEach message in the batch is processed\
        \ sequentially. The response contains a summary of successful and failed messages. Partial success is possible — some\
        \ messages may succeed while others fail.\n\n**Note on the `message` field:** In the POST request body, `message`\
        \ is a plain string (the text of the transcript). In the GET response, the same field name returns an array of content\
        \ objects (e.g. `[{\"type\":\"text\",\"text\":\"...\",\"isTextOnlyEmojis\":false}]`). Be aware of this type difference\
        \ when building integrations.\n\n**Author email resolution:** If `author.type` is `user` and `author.id` is an email\
        \ address, the API automatically resolves it to the corresponding `userId` for the account (best-effort — original\
        \ `author.id` is used if resolution fails).\n\n**410 Gone** is returned when the workflow session is found but has\
        \ already ended.\n\n## Finding Your Customer Subdomain\n\nYour subdomain is found in your Avaya Infinity portal URL\
        \ and is required for all API calls.\n\n**Example:** If your portal URL is:\n```\nhttps://core.avaya1234.ec.avayacloud.com/app/core-config-ui/\n\
        ```\nYour subdomain is: **`avaya1234`**\n\n**To use this API:**\n1. **Find your subdomain** from your Infinity portal\
        \ URL (as shown above)\n2. **Get your Bearer token** using your provisioned client credential (see Authentication\
        \ below)\n3. **In the API explorer on the right:**\n   - Click on `{customerId}` in the URL field and replace it with\
        \ your actual subdomain\n   - Paste your Bearer token in the Credentials section\n   - Fill out the request body with\
        \ your message batch\n\n## Authentication\n\nThis endpoint requires a valid OAuth 2.0 Bearer token. To get started,\
        \ contact Avaya Support to request a `client_id` and `client_secret` provisioned with the `TRANSCRIPT` scope.\n\n\
        Once you have your credentials, [generate an access token](https://developers.avayacloud.com/avaya-infinity/reference/access-token)\
        \ and include it in the `Authorization: Bearer <token>` header of each request.\n\n## Working with Interaction IDs\n\
        \nThe `{id}` path parameter accepts:\n- An **interactionId** for standard voice or digital interactions (format: `004d01...`)\n\
        - A **workflow session ID** for `WORKFLOW_SESSION` type sessions\n\nUse whichever ID corresponds to the interaction\
        \ context you are writing transcripts for.\n\n## Important Notes\n\n- Both `messages` array and `author.type` are\
        \ required. All other fields are optional.\n- Omitting `direction` defaults to `in` (inbound). Omitting `createdAt`\
        \ defaults to server time.\n- Partial success is normal — inspect `failed` and `errors` in the response even on a\
        \ `200`.\n- If `author.type` is `user` and `author.id` is an email, the API attempts to resolve it to the account's\
        \ `userId`. If resolution fails, the original `author.id` is preserved.\n- A `410 Gone` response indicates the workflow\
        \ session exists but has already ended — no new messages can be added.\n- Validation errors return `422 Unprocessable\
        \ Entity`, not `400`. Each error object includes `type`, `value`, `msg`, `path`, and `location`.\n\n## DOS Protection\
        \ & Rate Limiting\n- Implement retry logic with exponential backoff if you encounter rate limit errors.\n- Avoid sending\
        \ large batches in tight loops — space requests appropriately.\n\n## Security Implementation Guidelines\n- Never expose\
        \ your `client_secret` or Bearer token to client-side applications.\n- Implement proper error handling — avoid surfacing\
        \ internal error details to end users."
      parameters:
      - name: id
        in: path
        description: The interaction ID or workflow session ID to associate the transcript messages with. Accepts both interaction
          IDs and `WORKFLOW_SESSION` IDs.
        required: true
        schema:
          type: string
          example: 004d01000069d6b91753dcd4c0
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/PostTranscriptBatchRequest'
            examples:
              voice-transcript:
                summary: Voice call transcript batch
                value:
                  messages:
                  - message: Hello, I'm calling about my bill. There seems to be an error on last month's statement.
                    direction: in
                    author:
                      type: customer
                      id: customer_phone_session_123
                    languageCode: en-us
                    createdAt: '2024-10-17T10:30:56.789Z'
                  - message: Thank you for reaching out. I'll take a look at your billing history and get back to you shortly.
                    direction: out
                    author:
                      type: bot
                      id: billing_bot_001
                    languageCode: en-us
                    createdAt: '2024-10-17T10:31:10.123Z'
              minimal:
                summary: Minimal required fields
                value:
                  messages:
                  - message: Customer inquiry received.
                    author:
                      type: customer
                      id: cust-001
      responses:
        '200':
          description: Batch processed. Inspect `failed` and `errors` for partial failures.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/PostTranscriptBatchResponse'
              examples:
                allSucceeded:
                  summary: All messages processed successfully
                  value:
                    total: 2
                    successful: 2
                    failed: 0
                    results:
                    - index: 0
                      success: true
                      result:
                        status: 200
                        data:
                          message:
                            id: 049d011029452f5a95fa42e9a7
                    - index: 1
                      success: true
                      result:
                        status: 200
                        data:
                          message:
                            id: 049d0110290930ad042dc96ef6
                partialFailure:
                  summary: Partial failure — one message failed
                  value:
                    total: 2
                    successful: 1
                    failed: 1
                    results:
                    - index: 0
                      success: true
                      result:
                        status: 200
                        data:
                          message:
                            id: 049d011029452f5a95fa42e9a7
                    errors:
                    - index: 1
                      message: Thank you for reaching out. I will...
                      error: Channel not found
                      status: 404
        '401':
          description: 'Unauthorized — missing or invalid Bearer token. Note: At the gateway level, missing authentication
            may produce a `302` redirect rather than a JSON `401` response. If your client receives an unexpected redirect,
            check that your `Authorization` header is present and correctly formatted.'
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ErrorResponse'
              example:
                message: Unauthorized
        '404':
          description: Not Found — the interaction or workflow session does not exist.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ErrorResponse'
              example:
                message: Session not found
        '410':
          description: Gone — the workflow session was found but has already ended. No new transcript messages can be added.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ErrorResponse'
              example:
                message: Session has ended
        '422':
          description: Unprocessable Entity — one or more request fields failed validation. Each error object includes the
            field path, rejected value, and a description of the failure.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ValidationErrorResponse'
              examples:
                invalidAuthorType:
                  summary: Invalid author.type value
                  value:
                    message: Validation failed
                    errors:
                    - type: field
                      value: unknown
                      msg: author.type is required and must be "user", "bot", or "customer"
                      path: messages[0].author.type
                      location: body
                emptyMessages:
                  summary: Empty messages array
                  value:
                    message: Validation failed
                    errors:
                    - type: field
                      value: []
                      msg: messages must be a non-empty array
                      path: messages
                      location: body
        '500':
          description: Internal Server Error
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ErrorResponse'
              example:
                error: Internal server error during batch processing
                details: Unexpected failure
  /transcripts/{interactionId}:
    get:
      tags:
      - Transcripts
      summary: Get transcripts for an interaction
      operationId: transcripts.get
      security:
      - Bearer: []
      description: "Retrieves transcript messages for a given interaction or workflow session by its ID.\n\nThe `{interactionId}`\
        \ parameter accepts either an **interactionId** or a **workflow session ID**.\n\nReturns a sanitized subset of message\
        \ fields appropriate for transcript display. Internal and sensitive fields are excluded from the response.\n\n**Note\
        \ on the `message` field:** In the GET response, `message` is an array of content objects (e.g. `[{\"type\":\"text\"\
        ,\"text\":\"Hello\",\"isTextOnlyEmojis\":false}]`). This differs from the POST request body where `message` is a plain\
        \ string. Be aware of this type difference when building integrations.\n\n**Note on `createdAt` format:** The `createdAt`\
        \ field in responses uses the format `2026-02-26 16:27:05.124000` (space separator, no `T`, no timezone), rather than\
        \ standard ISO 8601. Parse accordingly.\n\n**Known issue:** If the `{interactionId}` does not exist, the API currently\
        \ returns a `500 Internal Server Error` (with `{\"message\":\"Internal Server Error\",\"errors\":[]}`) rather than\
        \ a `404`. This is a known behavior pending a server-side fix — treat a `500` response on this endpoint as a potential\
        \ \"not found\" condition.\n\n## Finding Your Customer Subdomain\n\nYour subdomain is found in your Avaya Infinity\
        \ portal URL and is required for all API calls.\n\n**Example:** If your portal URL is:\n```\nhttps://core.avaya1234.ec.avayacloud.com/app/core-config-ui/\n\
        ```\nYour subdomain is: **`avaya1234`**\n\n**To use this API:**\n1. **Find your subdomain** from your Infinity portal\
        \ URL (as shown above)\n2. **Get your Bearer token** using your provisioned client credential (see Authentication\
        \ below)\n3. **In the API explorer on the right:**\n   - Click on `{customerId}` in the URL field and replace it with\
        \ your actual subdomain\n   - Paste your Bearer token in the Credentials section\n   - Provide the `interactionId`\
        \ path parameter\n\n## Authentication\n\nThis endpoint requires a valid OAuth 2.0 Bearer token. To get started, contact\
        \ Avaya Support to request a `client_id` and `client_secret` provisioned with the `TRANSCRIPT` scope.\n\nOnce you\
        \ have your credentials, [generate an access token](https://developers.avayacloud.com/avaya-infinity/reference/access-token)\
        \ and include it in the `Authorization: Bearer <token>` header of each request.\n\n## Working with Interaction IDs\
        \ and Pagination\n\nThe `{interactionId}` path parameter accepts:\n- An **interactionId** for standard voice or digital\
        \ interactions (format: `004d01...`)\n- A **workflow session ID** for `WORKFLOW_SESSION` type sessions\n\nUse the\
        \ `before` and `after` query parameters with a `paginationId` or timestamp value to page through large transcript\
        \ histories. The `limit` parameter controls how many messages are returned per page.\n\n## Important Notes\n\n- Transcript\
        \ messages are returned in the order they were created.\n- The `message` field in each returned message is an array\
        \ of content objects, not a plain string.\n- The `createdAt` field format is `2026-02-26 16:27:05.124000` — not standard\
        \ ISO 8601. Parse using a space separator.\n- For customer authors, a root-level `displayName` field may be present\
        \ on the author object in addition to the `details.displayName`.\n- Messages posted via the Send Transcript Batch\
        \ endpoint will have `subType: \"transcription\"`.\n- **Known issue:** A non-existent `interactionId` returns `500`\
        \ instead of `404`. This is pending a server-side fix.\n\n## DOS Protection & Rate Limiting\n- Implement retry logic\
        \ with exponential backoff for transient errors.\n- Use pagination parameters (`limit`, `before`, `after`) to retrieve\
        \ large transcripts in manageable pages rather than fetching all messages in one call.\n\n## Security Implementation\
        \ Guidelines\n- Never expose your `client_secret` or Bearer token to client-side applications.\n- Implement proper\
        \ error handling — avoid surfacing internal error details to end users."
      parameters:
      - name: interactionId
        in: path
        description: The interaction ID or workflow session ID whose transcripts should be retrieved. Accepts both interaction
          IDs and `WORKFLOW_SESSION` IDs.
        required: true
        schema:
          type: string
          example: 004d01000069d6b91753dcd4c0
      - name: limit
        in: query
        description: Maximum number of messages to return per page.
        required: false
        schema:
          type: integer
          example: 50
      - name: before
        in: query
        description: Return messages created before this pagination cursor (`paginationId` or timestamp). Used for backward
          pagination through results.
        required: false
        schema:
          type: string
          example: pag-001
      - name: after
        in: query
        description: Return messages created after this pagination cursor (`paginationId` or timestamp). Used for forward
          pagination through results.
        required: false
        schema:
          type: string
          example: pag-001
      - name: type
        in: query
        description: Filter messages by message type (e.g. `message`, `event`).
        required: false
        schema:
          type: string
          example: message
      responses:
        '200':
          description: OK — returns transcript messages for the interaction.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/GetTranscriptsResponse'
              examples:
                withMessages:
                  summary: Interaction with messages
                  value:
                    messages:
                    - id: 049d011029452f5a95fa42e9a7
                      channel: 004d01000069d6b91753dcd4c0
                      direction: out
                      type: message
                      subType: null
                      paginationId: pag-001
                      author:
                        type: user
                        details:
                          id: 0020005ca02e4654dfde78
                          displayName: Agent Smith
                          email: agent@example.com
                          title: Support Agent
                      message:
                      - type: text
                        text: Hello, how can I help you today?
                        isTextOnlyEmojis: false
                      languageCode: en-US
                      createdAt: '2026-02-26 10:30:00.000000'
                    - id: 049d0110290930ad042dc96ef6
                      channel: 004d01000069d6b91753dcd4c0
                      direction: in
                      type: message
                      subType: transcription
                      paginationId: pag-002
                      author:
                        type: customer
                        displayName: John Doe
                        details:
                          id: cust-12345
                          displayName: John Doe
                          email: null
                          title: null
                      message:
                      - type: text
                        text: I need help with my account.
                        isTextOnlyEmojis: false
                      languageCode: en-US
                      createdAt: '2026-02-26 10:30:15.000000'
                emptyTranscript:
                  summary: No messages found
                  value:
                    messages: []
        '401':
          description: 'Unauthorized — missing or invalid Bearer token. Note: At the gateway level, missing authentication
            may produce a `302` redirect rather than a JSON `401` response. If your client receives an unexpected redirect,
            check that your `Authorization` header is present and correctly formatted.'
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ErrorResponse'
              example:
                message: Unauthorized
        '500':
          description: Internal Server Error. **Known issue:** When the `{interactionId}` does not exist, the API currently
            returns `500` with `{"message":"Internal Server Error","errors":[]}` rather than a `404`. Treat this response
            as a potential "not found" condition when the interaction ID has not been verified to exist.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/ErrorResponse'
              example:
                message: Internal Server Error
                errors: []
components:
  securitySchemes:
    Bearer:
      type: http
      scheme: bearer
      description: 'To use this API, contact Avaya Support to request a `client_id` and `client_secret` provisioned with the
        `TRANSCRIPT` scope.


        Once you have your credentials, [generate an access token](https://developers.avayacloud.com/avaya-infinity/reference/access-token)
        and include it in the `Authorization: Bearer <token>` header of each request.'
  schemas:
    Author:
      title: Author
      type: object
      description: The author of a transcript message in a POST request body.
      required:
      - type
      - id
      properties:
        type:
          type: string
          description: The role of the author. Valid values are `user` (agent), `bot`, or `customer`.
          enum:
          - user
          - bot
          - customer
          example: customer
        id:
          type: string
          description: The unique identifier of the author. For `user` type, this can be a user ID or email address — if an
            email is provided, the API will attempt to resolve it to the corresponding `userId`.
          example: customer_phone_session_123
        displayName:
          type: string
          description: The display name of the author. Optional.
          example: John Doe
        name:
          type: string
          description: The name of the author. Optional.
          example: John Doe
    BatchMessage:
      title: BatchMessage
      type: object
      description: A single transcript message to be sent in a POST batch request.
      required:
      - message
      - author
      properties:
        message:
          type: string
          description: The text content of the transcript message. Note that in the GET response, the field of the same name
            returns an array of content objects rather than a plain string.
          example: Hello, I'm calling about my bill.
        direction:
          type: string
          description: The direction of the message — `in` for inbound (customer) or `out` for outbound (agent/bot). Defaults
            to `in` if omitted.
          enum:
          - in
          - out
          example: in
        author:
          $ref: '#/components/schemas/Author'
        languageCode:
          type: string
          description: BCP-47 language code for the message. Optional.
          example: en-us
        createdAt:
          type: string
          format: date-time
          description: ISO 8601 timestamp indicating when the message was created. Optional — defaults to server time if omitted.
          example: '2024-10-17T10:30:56.789Z'
    PostTranscriptBatchRequest:
      title: PostTranscriptBatchRequest
      type: object
      description: Request body for sending a batch of transcript messages.
      required:
      - messages
      properties:
        messages:
          type: array
          description: A non-empty array of transcript messages to send. Each message is processed sequentially.
          minItems: 1
          items:
            $ref: '#/components/schemas/BatchMessage'
    MessageProcessResult:
      title: MessageProcessResult
      type: object
      description: Result for a successfully processed message in the batch.
      properties:
        index:
          type: integer
          description: The zero-based index of the message in the request array.
          example: 0
        success:
          type: boolean
          description: Indicates successful processing.
          example: true
        result:
          type: object
          properties:
            status:
              type: integer
              description: HTTP status code of the underlying channel message API call.
              example: 200
            data:
              type: object
              description: The created message data.
              properties:
                message:
                  type: object
                  description: The created channel message object.
                  properties:
                    id:
                      type: string
                      description: The ID of the newly created channel message.
                      example: 049d011029452f5a95fa42e9a7
    MessageProcessError:
      title: MessageProcessError
      type: object
      description: Error detail for a message that failed to process in the batch.
      properties:
        index:
          type: integer
          description: The zero-based index of the message in the request array.
          example: 1
        message:
          type: string
          description: A truncated preview of the message content (up to 50 characters).
          example: I need help with my...
        error:
          type: string
          description: The error message.
          example: Channel not found
        status:
          type: integer
          description: HTTP status code associated with the error.
          example: 404
        details:
          description: Additional error details. Optional.
    PostTranscriptBatchResponse:
      title: PostTranscriptBatchResponse
      type: object
      description: Response summarising the result of a transcript batch submission.
      properties:
        total:
          type: integer
          description: Total number of messages submitted in the request.
          example: 2
        successful:
          type: integer
          description: Number of messages successfully processed.
          example: 2
        failed:
          type: integer
          description: Number of messages that failed to process.
          example: 0
        results:
          type: array
          description: Array of results for successfully processed messages.
          items:
            $ref: '#/components/schemas/MessageProcessResult'
        errors:
          type: array
          description: Array of errors for messages that failed. Omitted when all messages succeed.
          items:
            $ref: '#/components/schemas/MessageProcessError'
    MessageContentItem:
      title: MessageContentItem
      type: object
      description: A single content component within a message as returned by the GET endpoint.
      properties:
        type:
          type: string
          description: The content type of this component (e.g. `text`).
          example: text
        text:
          type: string
          description: The text content of this component.
          example: Hello, how can I help you today?
        isTextOnlyEmojis:
          type: boolean
          description: Indicates whether the text content consists entirely of emoji characters.
          example: false
    TranscriptMessageAuthorDetails:
      title: TranscriptMessageAuthorDetails
      type: object
      description: Detailed information about the message author, as returned in the GET response.
      properties:
        id:
          type: string
          nullable: true
          description: The author's unique identifier.
          example: 0020005ca02e4654dfde78
        displayName:
          type: string
          nullable: true
          description: The author's display name.
          example: Agent Smith
        email:
          type: string
          nullable: true
          description: The author's email address.
          example: agent@example.com
        title:
          type: string
          nullable: true
          description: The author's title or role.
          example: Support Agent
    TranscriptMessageAuthor:
      title: TranscriptMessageAuthor
      type: object
      description: Author information attached to a retrieved transcript message in the GET response.
      properties:
        type:
          type: string
          description: The role of the author. In GET responses, this may include `email` or `phoneNumber` in addition to
            `user` and `customer`.
          enum:
          - user
          - customer
          - email
          - phoneNumber
          example: user
        displayName:
          type: string
          nullable: true
          description: Root-level display name of the author. Present on customer author objects — may appear outside the
            `details` object.
          example: John Doe
        details:
          $ref: '#/components/schemas/TranscriptMessageAuthorDetails'
    TranscriptMessage:
      title: TranscriptMessage
      type: object
      description: 'A transcript message returned by the GET endpoint. Contains a sanitized subset of channel message fields
        — internal and sensitive fields are excluded.


        Note that the `message` field here is an array of content objects, not a plain string. This differs from the POST
        request body where `message` is a string.'
      properties:
        id:
          type: string
          description: Unique identifier for the channel message.
          example: 049d011029452f5a95fa42e9a7
        channel:
          type: string
          description: The channel (interaction) ID this message belongs to.
          example: 004d01000069d6b91753dcd4c0
        author:
          $ref: '#/components/schemas/TranscriptMessageAuthor'
        type:
          type: string
          description: The message type (e.g. `message`, `event`).
          example: message
        subType:
          type: string
          nullable: true
          description: Optional sub-type of the message. Messages posted via the Send Transcript Batch endpoint will have
            `subType` set to `"transcription"`.
          example: transcription
        direction:
          type: string
          description: Message direction — `in` (inbound from customer) or `out` (outbound from agent or bot).
          enum:
          - in
          - out
          example: out
        paginationId:
          type: string
          description: Opaque cursor value used to paginate through messages using the `before` or `after` query parameters.
          example: pag-001
        message:
          type: array
          description: The message content represented as an array of content components. Each item contains a `type`, `text`,
            and `isTextOnlyEmojis` field.
          items:
            $ref: '#/components/schemas/MessageContentItem'
        languageCode:
          type: string
          description: BCP-47 language code of the message.
          example: en-US
        createdAt:
          type: string
          description: Timestamp of when the message was created. **Note:** The format returned is `2026-02-26 16:27:05.124000`
            (space separator, no `T`, no timezone) rather than standard ISO 8601. Parse using a space as the date-time separator.
          example: '2026-02-26 16:27:05.124000'
    GetTranscriptsResponse:
      title: GetTranscriptsResponse
      type: object
      description: Response containing transcript messages for an interaction.
      properties:
        messages:
          type: array
          description: Array of transcript messages for the interaction. Returns an empty array if no messages exist.
          items:
            $ref: '#/components/schemas/TranscriptMessage'
    ValidationErrorItem:
      title: ValidationErrorItem
      type: object
      description: A single validation error returned in a 422 response.
      properties:
        type:
          type: string
          description: The category of validation error (e.g. `field`).
          example: field
        value:
          description: The rejected value that caused the validation failure.
          example: unknown
        msg:
          type: string
          description: Human-readable description of the validation failure.
          example: author.type is required and must be "user", "bot", or "customer"
        path:
          type: string
          description: The dot-notation path to the field that failed validation.
          example: messages[0].author.type
        location:
          type: string
          description: Where in the request the field was found (e.g. `body`).
          example: body
    ValidationErrorResponse:
      title: ValidationErrorResponse
      type: object
      description: Response body for 422 Unprocessable Entity validation errors.
      properties:
        message:
          type: string
          description: High-level description of the failure.
          example: Validation failed
        errors:
          type: array
          description: Array of individual validation errors.
          items:
            $ref: '#/components/schemas/ValidationErrorItem'
    ErrorResponse:
      title: ErrorResponse
      type: object
      description: Standard error response.
      properties:
        message:
          type: string
          description: Human-readable error message.
          example: Not found
        error:
          type: string
          description: Error identifier or short description.
          example: id parameter is required (interactionId or workflow session id)
        errors:
          type: array
          description: Additional error detail array. May be empty.
          items:
            type: object