arazzo: 1.0.1
info:
  title: Claude Multi-Turn Conversation
  summary: Send a first user turn, then continue the conversation with a follow-up that includes the assistant's reply.
  description: >-
    A two-turn conversation flow. The workflow sends an opening user message,
    captures the assistant's text reply, and then issues a second request that
    replays the full conversation — the original user turn, the assistant
    response, and a new follow-up user turn — so the model has the prior context.
    This demonstrates how the stateless Messages API is chained into a
    multi-turn dialogue. Every step spells out its request inline — including
    the required x-api-key and anthropic-version headers — so the flow can be
    read and executed without opening the underlying OpenAPI description.
  version: 1.0.0
sourceDescriptions:
- name: claudeApi
  url: ../openapi/claude-messages-api.yml
  type: openapi
workflows:
- workflowId: multi-turn-conversation
  summary: Chain two Messages API calls into a single multi-turn conversation.
  description: >-
    Sends an opening user prompt, then sends a follow-up prompt that includes
    the prior user and assistant turns so the model continues with full context.
  inputs:
    type: object
    required:
    - apiKey
    - model
    - firstPrompt
    - followUpPrompt
    properties:
      apiKey:
        type: string
        description: Anthropic API key sent in the x-api-key header.
      anthropicVersion:
        type: string
        description: Value for the required anthropic-version header.
        default: '2023-06-01'
      model:
        type: string
        description: The model to use for both turns of the conversation.
      firstPrompt:
        type: string
        description: The opening user message.
      followUpPrompt:
        type: string
        description: The follow-up user message sent after the assistant replies.
      maxTokens:
        type: integer
        description: Maximum number of tokens to generate per turn.
        default: 1024
  steps:
  - stepId: openingTurn
    description: >-
      Send the opening user message and capture the assistant's reply content
      so it can be replayed as context in the next turn.
    operationId: createMessage
    parameters:
    - name: x-api-key
      in: header
      value: $inputs.apiKey
    - name: anthropic-version
      in: header
      value: $inputs.anthropicVersion
    requestBody:
      contentType: application/json
      payload:
        model: $inputs.model
        max_tokens: $inputs.maxTokens
        messages:
        - role: user
          content: $inputs.firstPrompt
    successCriteria:
    - condition: $statusCode == 200
    outputs:
      firstMessageId: $response.body#/id
      assistantContent: $response.body#/content
      stopReason: $response.body#/stop_reason
  - stepId: followUpTurn
    description: >-
      Send the follow-up user message with the prior user turn and assistant
      reply included so the model continues the conversation with full context.
    operationId: createMessage
    parameters:
    - name: x-api-key
      in: header
      value: $inputs.apiKey
    - name: anthropic-version
      in: header
      value: $inputs.anthropicVersion
    requestBody:
      contentType: application/json
      payload:
        model: $inputs.model
        max_tokens: $inputs.maxTokens
        messages:
        - role: user
          content: $inputs.firstPrompt
        - role: assistant
          content: $steps.openingTurn.outputs.assistantContent
        - role: user
          content: $inputs.followUpPrompt
    successCriteria:
    - condition: $statusCode == 200
    outputs:
      followUpMessageId: $response.body#/id
      content: $response.body#/content
      stopReason: $response.body#/stop_reason
      usage: $response.body#/usage
  outputs:
    firstMessageId: $steps.openingTurn.outputs.firstMessageId
    followUpMessageId: $steps.followUpTurn.outputs.followUpMessageId
    content: $steps.followUpTurn.outputs.content
    usage: $steps.followUpTurn.outputs.usage