Connect WorkBuddy
Add a custom text model in WorkBuddy's model editor and use the correct Chat Completions endpoint behavior.
https://aelder.dev/v1Use an existing WorkBuddy desktop installation with its visual model editor. Prepare a service key and exact model ID from our models endpoint.
Add a custom model
Open Settings → Model, add a model and select Custom as the provider.
| Field | Value |
|---|---|
| URL | The service Base URL above, including /v1 |
| API Key | Your service key, without Bearer |
| Model name | The exact service model ID |
| Custom Protocol | Off |
With Custom Protocol off, WorkBuddy completes the standard /chat/completions path. Turning it on sends requests to the entered URL as-is, so a Base URL ending only in /v1 would be incomplete.
Set context/output limits from verified information for your chosen model. Enable tool calling only for a compatible model and keep image input off for this text setup. Click Save, then select the custom model in the chat model picker instead of Auto Mode. These controls and their persistence are documented in WorkBuddy’s model guide.
Verify the connection
Start a new task:
Reply with CONNECTION_OK. Do not use tools or modify files.Compare the reply with the matching service request record. Then try a small file-reading task with normal approvals if you need agent tools. Each request uses API balance.
A saved model is not a successful request. For a 404, check Custom Protocol and the URL; for a missing model, check the exact ID and key permissions. If tools fail after text succeeds, check model capabilities. See our API troubleshooting.
To switch back, choose your previous model. WorkBuddy’s model guide describes a legacy file location but does not establish a current file-editing schema; use the supported editor for this setup.
Need the key? Open your AELDER dashboard. For direct API calls, see the API reference.

