BrokenGPT Node.js API quickstart

1. Create a scoped API key in the BrokenGPT dashboard.
2. Install the client with npm install openai.
3. Put BROKENGPT_API_KEY in your host secret store or local .env file that is excluded from Git.

npm install openai

# .env.local (never commit this value)
BROKENGPT_API_KEY=your-key

Run this code only on your server, worker, or trusted command line. The base URL is https://brokengpt.com/v1 and the public model alias is broken-one.

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.BROKENGPT_API_KEY,
  baseURL: "https://brokengpt.com/v1",
});

const response = await client.chat.completions.create({
  model: "broken-one",
  messages: [{ role: "user", content: "Give me three test cases." }],
});

console.log(response.choices[0]?.message?.content);

In an HTTP application, propagate caller cancellation and avoid assuming every stream ends normally. Keep partially generated content separate from committed application state until your completion policy accepts it.

const stream = await client.chat.completions.create({
  model: "broken-one",
  messages: [{ role: "user", content: "Draft a concise release note." }],
  stream: true,
});

for await (const chunk of stream) {
  const text = chunk.choices[0]?.delta?.content;
  if (text) process.stdout.write(text);
}

• Set an application timeout and abort requests that no longer have a caller.
• Handle 400, 401, 403, 404, 429, and 5xx responses by category.
• Use bounded exponential backoff only for retryable responses.
• Validate JSON and tool arguments before they reach databases, shells, or external APIs.
• Track latency, usage, and request IDs without logging credentials.