@prompty/openai

OpenAI provider for Prompty — executor and processor for the OpenAI API.

Installation

npm install @prompty/core @prompty/openai openai

Usage

Import @prompty/openai to auto-register the openai provider, then use @prompty/core as normal:

import "@prompty/openai";
import { run } from "@prompty/core";

const result = await run("./chat.prompty", {
  question: "What is quantum computing?",
});

.prompty Configuration

Set provider: openai in your .prompty file:

---
name: my-prompt
model:
  id: gpt-4o-mini
  provider: openai
  apiType: chat
  connection:
    kind: key
    endpoint: ${env:OPENAI_BASE_URL}
    apiKey: ${env:OPENAI_API_KEY}
  options:
    temperature: 0.7
    maxOutputTokens: 1000
---
system:
You are a helpful assistant.

user:
{{question}}

Supported API Types

| apiType | Description | |-----------|-------------| | chat (default) | Chat completions via client.chat.completions.create() | | embedding | Embeddings via client.embeddings.create() | | image | Image generation via client.images.generate() | | responses | Responses API via client.responses.create() |

Streaming

Enable streaming via additionalProperties:

model:
  options:
    additionalProperties:
      stream: true

Returns a PromptyStream that can be iterated asynchronously.

Exports

| Export | Description | |--------|-------------| | OpenAIExecutor | Executor implementation for OpenAI | | OpenAIProcessor | Processor for OpenAI responses | | processResponse | Shared response processing helper | | messageToWire | Convert Message → OpenAI wire format | | buildChatArgs | Build chat completion arguments | | buildEmbeddingArgs | Build embedding arguments | | buildImageArgs | Build image generation arguments | | buildResponsesArgs | Build Responses API arguments |

License

MIT