Skip to content

BrokenCode connection

Connect BrokenCode to BrokenGPT

Connect BrokenCode to your BrokenGPT account for privately routed open-model inference, with a clear local credential flow.

UPDATED 03 Aug 20263 min readTECHNICAL GUIDE
01

BrokenGPT is the supported hosted connection.

Select BrokenGPT in the connection panel and complete authorization in the browser. BrokenCode stores the resulting credential locally and uses the stable broken-one model alias.

  1. Open Settings, then Model connections.
  2. Choose Connect BrokenGPT.
  3. Authorize the device in your browser and return to BrokenCode.
  4. Confirm the connection indicator is green before starting a session.
02

Credentials stay out of project files.

  • Never paste a token into brokencode.json or commit it to a repository.
  • Disconnect from Settings to revoke the local BrokenCode link.
  • Reconnect if authorization was interrupted or the local credential expired.
03

Custom model providers are coming soon.

The settings surface shows upcoming provider options without sending requests through an incomplete connection. Use BrokenGPT for the live hosted path today.

STRAIGHT ANSWERS

Frequently asked questions

01What if no connection is configured?

BrokenCode asks you to connect BrokenGPT. It does not substitute a placeholder response.

02Are custom providers available?

They are listed as coming soon. BrokenGPT is the current supported hosted connection.

Connect

Link BrokenCode to BrokenGPT.

Sign in once, authorize the device, and keep inference configuration outside the repository.

Authorize a device