arazzo: 1.0.1
info:
  title: OpenAI Create Embedding
  summary: Confirm an embedding model exists, then embed input text.
  description: >-
    Lists the available models to confirm the requested embedding model is
    present, then submits input text to the embeddings endpoint and returns the
    resulting vector. Every step spells out its request inline so the flow can
    be read and executed without opening the underlying OpenAPI description.
  version: 1.0.0
sourceDescriptions:
- name: openaiApi
  url: ../openapi/openai-openapi-master.yml
  type: openapi
workflows:
- workflowId: create-embedding
  summary: Create an embedding vector for input text.
  description: >-
    Verifies the embedding model is available, then sends input text to the
    embeddings endpoint and returns the generated embedding vector and token
    usage.
  inputs:
    type: object
    required:
    - apiKey
    - model
    - input
    properties:
      apiKey:
        type: string
        description: OpenAI API key used as a Bearer token.
      model:
        type: string
        description: The embedding model id (e.g. text-embedding-3-small).
      input:
        type: string
        description: The text to turn into an embedding vector.
  steps:
  - stepId: listModels
    description: List available models to confirm the embedding model is present.
    operationId: listModels
    parameters:
    - name: Authorization
      in: header
      value: "Bearer $inputs.apiKey"
    successCriteria:
    - condition: $statusCode == 200
    outputs:
      models: $response.body#/data
  - stepId: createEmbedding
    description: Create an embedding vector for the supplied input text.
    operationId: createEmbedding
    parameters:
    - name: Authorization
      in: header
      value: "Bearer $inputs.apiKey"
    requestBody:
      contentType: application/json
      payload:
        model: $inputs.model
        input: $inputs.input
        encoding_format: float
    successCriteria:
    - condition: $statusCode == 200
    outputs:
      embedding: $response.body#/data/0/embedding
      promptTokens: $response.body#/usage/prompt_tokens
  outputs:
    embedding: $steps.createEmbedding.outputs.embedding
    promptTokens: $steps.createEmbedding.outputs.promptTokens