llm-prompt-stream

📜 llm-prompt-stream

A lightweight Node.js module for streaming OpenAI responses with Markdown formatting and file saving support.

🚀 Features

• ✅ Streaming OpenAI responses in real-time.
• ✅ Markdown support for structured responses.
• ✅ Write streamed responses to a file.
• ✅ Mock data support for testing.

📦 Installation

Install via npm:

npm install llm-prompt-stream

🔧 Usage

Basic Example: Streaming an OpenAI Completion

import { createCompletionAndStream, readStream } from "llm-prompt-stream";

// Your OpenAI API key
const openAIKey = "your-api-key";

const messages = [
  { role: "system", content: "You are a helpful assistant." },
  { role: "user", content: "Tell me a fun fact!" },
];

async function run() {
  const stream = await createCompletionAndStream(openAIKey, messages);
  const response = await readStream(stream);
  console.log("Full Response:", response);
}

run();

Saving Streamed Response to a Markdown File

import { createCompletionAndStream, readStream } from "llm-prompt-stream";

const openAIKey = "your-api-key";
const messages = [{ role: "user", content: "Give me a summary of AI." }];

async function run() {
  const stream = await createCompletionAndStream(openAIKey, messages);
  await readStream(stream, true, "output.md");
  console.log("Response saved to output.md!");
}

run();

🛠 API Reference

`createCompletionAndStream(openAIKey: string, messages: any[])`

Creates an OpenAI streaming completion.

• `openAIKey` - Your OpenAI API key.
• `messages` - Array of message objects for the chat completion.

Returns: A `ReadableStream` containing the OpenAI response.

`readStream(stream: ReadableStream, createFile?: boolean, outputFilename?: string)`

Reads and processes a streamed response.

• `stream` - The OpenAI response stream.
• `createFile` (optional) - If `true`, writes response to a file.
• `outputFilename` (optional) - Filename for the saved response (default: `response.md`).

Returns: A string containing the full response.

🧪 Running Tests

To run tests locally:

npm test

📜 License

This project is licensed under the MIT License.

🤝 Contributing

Contributions are welcome! Feel free to open an issue or submit a PR.

• Fork the repo
• Clone it: `git clone https://github.com/yourusername/llm-prompt-stream.git\`
• Install dependencies: `npm install`
• Make changes & commit
• Submit a pull request!

⭐ Support

If you like this package, give it a ⭐ on GitHub!