---
title: "Connect to Auto"
method: GET
path: "/v1/org/auto-router/connect"
tags: ["Inference Providers"]
---

# Connect to Auto

`GET /v1/org/auto-router/connect`

## Request

- Base URL: `https://api.openworklabs.com`
- URL: `https://api.openworklabs.com/v1/org/auto-router/connect`
- Auth: one of:
  - API key in header `x-api-key`
  - HTTP bearer

For desktops: when the organization's Auto Router is enabled and the caller can use at least one of its models, returns the OpenAI-compatible gateway base URL and the caller's own Gateway key so the desktop can offer Auto. Answers 404 otherwise, including while the AI Gateway has no TypeSafe key. Never returns provider secrets. Requires organization membership and the Auto Router feature.

## Response `200`

Connection returned.

- AutoRouterConnect
  - `model` object, required
    - `id` 'openwork-auto', required
    - `name` string, required
  - `baseURL` string, required
  - `apiKey` string, required

## Other responses

- `401`: Authentication required.
- `404`: The Auto Router is off, not configured, or none of its models are available to the caller.

## Changes

- **2026-10-27** `7e95da81abe4`: 1 info
  - endpoint added

[Operation history](https://skmtc.dev/openworklabs/apis/den-api/changes/v1/org/auto-router/connect/get.md)

---

[API](https://skmtc.dev/openworklabs/apis/den-api.md) - [All operations](https://skmtc.dev/openworklabs/apis/den-api/llms.txt) - [OpenAPI document](https://skmtc.dev/openworklabs/apis/den-api/revisions/8117bc7c5626?raw)
