APIMaster.ai
APIMaster.ai Docs
Quick startAPIMaster.aiMarketplace

WorkBuddy Third-Party Key Setup — APIMaster.ai Custom Models Guide

How to add an APIMaster.ai OpenAI-compatible API key in WorkBuddy. Open Settings → Models → Add Model, set Endpoint, API Key, and Model Name, then select the custom model in chat.

WorkBuddy is a desktop AI agent (workspace files, skills/commands, multi-model switching). APIMaster.ai provides an OpenAI-compatible API — add a Custom Model in WorkBuddy to connect.

Get an API Key first. Keys are stored locally in %USERPROFILE%\.workbuddy\models.json (Windows) or ~/.workbuddy/models.json (macOS / Linux). Do not share them in screenshots or chat.

Prerequisites

  1. WorkBuddy desktop app installed and running.
  2. API Key copied from the APIMaster console.
  3. Target model id from the marketplace (e.g. gpt-5.6-sol, claude-sonnet-4-6, deepseek-v4-pro).

Step 1 — Open custom model settings

Use either entry point:

Option A (recommended) — from the chat model picker:

  1. Next to the input box, click the current model name (e.g. Auto).
  2. At the bottom of the menu, click Configure custom models.

WorkBuddy model picker → Configure custom models

Option B — from Settings:

  1. Open Settings.
  2. Select Models in the sidebar.
  3. Under CUSTOM MODELS, click + Add Model.

Settings → Models → + Add Model

Local config file:

  • Windows: %USERPROFILE%\.workbuddy\models.json
  • macOS / Linux: ~/.workbuddy/models.json

Step 2 — Fill in APIMaster connection details

In the Add Model dialog (OpenAI-compatible API required):

Field Value
Provider Custom
Endpoint https://apimaster.ai/v1 (/v1 required)
API Key Your APIMaster key
Model Name Exact model id from the marketplace, e.g. gpt-5.6-sol, claude-sonnet-4-6

Advanced Settings (recommended):

Option Suggestion
Tool Calling ✅ Enable (needed for agent tools / skills)
Image Input ✅ Enable if the model supports vision
Reasoning Optional — leave off for most chat models
Custom Protocol ❌ Leave off (standard OpenAI-compatible)

Leave Input / Output on Use provider default, or pick presets (e.g. Input 128K, Output 16K).

Add Model: Endpoint = apimaster.ai/v1 + Model Name

Click Save.

Notes:

  • Model Name must match the marketplace id exactly (case and hyphens).
  • To use another model, + Add Model again — reuse the same Endpoint and Key.
  • Saved models appear under SAVED MODELS and in the chat picker’s Custom Models section.

Step 3 — Select the model in chat

  1. Open the model picker on the main screen.
  2. Under Custom Models, select the model you added (e.g. gpt-5.6-sol).
  3. Send a test message (e.g. hi).
  4. The reply should show the model name next to the action bar — traffic is going through APIMaster.

Successful chat with model label gpt-5.6-sol

Example model ids

Aligned with APIMaster GET /v1/models (availability depends on your account):

model id Notes
gpt-5.6-sol GPT example used in screenshots
gpt-5.4 / gpt-5.5 GPT family
claude-sonnet-4-6 Claude Sonnet
claude-opus-4-8 Claude Opus
deepseek-v4-pro / deepseek-v4-flash DeepSeek
minimax-m3 / kimi-k3 Other popular models

Copy ids from the marketplace — do not guess aliases.

Troubleshooting

Issue Fix
401 / Unauthorized Re-paste API Key and Save
404 / model not found Re-copy Model Name from marketplace
Connection timeout Endpoint must be https://apimaster.ai/v1; check network/proxy
Empty Custom Models list Confirm Save; restart WorkBuddy
Tool calls fail Enable Tool Calling in Advanced Settings
Can't find settings Configure custom models in picker, or Settings → Models → + Add Model

Checklist

  • WorkBuddy installed
  • Provider = Custom
  • Endpoint = https://apimaster.ai/v1
  • Model Name from marketplace
  • Tool Calling enabled if needed; clicked Save
  • Model selectable under Custom Models and replies work

Related