arazzo: 1.0.1
info:
  title: Vercel AI Gateway Embed Then Summarize
  summary: Create an embedding for a document, then run a chat completion that summarizes it.
  description: >-
    A two-operation flow that combines the Vercel AI Gateway embedding and chat
    endpoints. The workflow first embeds the supplied document text so the vector
    is available for indexing or downstream retrieval, then issues a chat
    completion that asks a model to summarize the same text. This is a common
    retrieval-and-generate building block. Each step inlines its request and
    documents the success status.
  version: 1.0.0
sourceDescriptions:
- name: aiGatewayApi
  url: ../openapi/vercel-ai-gateway-openapi.yml
  type: openapi
workflows:
- workflowId: embedThenSummarize
  summary: Embed a document, then summarize it with a chat completion.
  description: >-
    Creates an embedding for the document text and then asks a chat model to
    summarize that same text, returning both the embedding and the summary.
  inputs:
    type: object
    required:
    - apiKey
    - embeddingModel
    - chatModel
    - document
    properties:
      apiKey:
        type: string
        description: Vercel AI Gateway API key (AI_GATEWAY_API_KEY) used as a Bearer token.
      embeddingModel:
        type: string
        description: "Embedding model id (e.g. openai/text-embedding-3-small)."
      chatModel:
        type: string
        description: "Chat model id in 'provider/model-name' form."
      document:
        type: string
        description: The document text to embed and summarize.
  steps:
  - stepId: embed
    description: Create an embedding for the document text.
    operationId: createEmbedding
    parameters:
    - name: Authorization
      in: header
      value: Bearer $inputs.apiKey
    requestBody:
      contentType: application/json
      payload:
        model: $inputs.embeddingModel
        input: $inputs.document
        encoding_format: float
    successCriteria:
    - condition: $statusCode == 200
    outputs:
      embedding: $response.body#/data/0/embedding
  - stepId: summarize
    description: Ask a chat model to summarize the same document text.
    operationId: createChatCompletion
    parameters:
    - name: Authorization
      in: header
      value: Bearer $inputs.apiKey
    requestBody:
      contentType: application/json
      payload:
        model: $inputs.chatModel
        messages:
        - role: system
          content: You are a concise summarizer. Summarize the user's document in three sentences.
        - role: user
          content: $inputs.document
        stream: false
    successCriteria:
    - condition: $statusCode == 200
    outputs:
      summary: $response.body#/choices/0/message/content
      totalTokens: $response.body#/usage/total_tokens
  outputs:
    embedding: $steps.embed.outputs.embedding
    summary: $steps.summarize.outputs.summary
    totalTokens: $steps.summarize.outputs.totalTokens