WorkBuddy

Prepare API Access

Have the three items below ready before you configure anything:

API Key

The key string you generated when creating your API Key.

Base URL

https://api.watt-cloud.com/v1

The trailing /v1 is required — do not drop it.

Model Name

Take the complete model identifier from the Model Marketplace, for example DeepSeek-V4-Pro.

Installing WorkBuddy / CodeBuddy

Grab the installer from the CodeBuddy official website and set it up on your machine.

Apart from the name, nothing separates WorkBuddy from CodeBuddy — features and setup are the same.

• WorkBuddy reads its config from ~/.workbuddy/models.json
• CodeBuddy reads its config from ~/.codebuddy/models.json

Connect WorkBuddy / CodeBuddy to Watt Cloud

Edit the Configuration File

Custom models are registered in the models.json file, which can live at two different levels:

• Project-level: <project-root>/.codebuddy/models.json
• User-level (global): ~/.codebuddy/models.json

When both files exist, the project-level one wins: any model ID defined there overrides the same ID from user-level configuration.

Sample Configuration

Open ~/.codebuddy/models.json and insert a block like the one below:

{
  "models": [
    {
      "id": "glm-5.1",
      "name": "GLM-5.1",
      "vendor": "watt",
      "apiKey": "your-api-key",
      "maxInputTokens": 200000,
      "maxOutputTokens": 8192,
      "url": "https://api.watt-cloud.com/v1/chat/completions",
      "supportsToolCall": true,
      "supportsImages": true
    },
    {
      "id": "deepseek-v4-pro",
      "name": "DeepSeek-V4-Pro",
      "vendor": "watt",
      "apiKey": "your-api-key",
      "maxInputTokens": 200000,
      "maxOutputTokens": 8192,
      "url": "https://api.watt-cloud.com/v1/chat/completions",
      "supportsToolCall": true,
      "supportsImages": false
    },
    {
      "id": "minimax-m2.7",
      "name": "MiniMax-M2.7",
      "vendor": "watt",
      "apiKey": "your-api-key",
      "maxInputTokens": 200000,
      "maxOutputTokens": 8192,
      "url": "https://api.watt-cloud.com/v1/chat/completions",
      "supportsToolCall": true,
      "supportsImages": false
    }
  ],
  "availableModels": [
    "glm-5.1",
    "deepseek-v4-pro",
    "minimax-m2.7"
  ]
}

Field Reference

Field	Description
id	A custom, unique identifier for the model — it labels the model for display and is what availableModels entries point to
name	Human-readable display label
vendor	A custom label naming the provider
apiKey	The API Key you created
url	The complete endpoint path — it has to end with /chat/completions
maxInputTokens	Upper bound on input tokens
maxOutputTokens	Upper bound on output tokens
supportsToolCall	Whether the model supports tool calling
supportsImages	Whether the model accepts image input
availableModels	Limits the model dropdown to the listed IDs; leave it empty to show everything

Warning: point url at the full path https://api.watt-cloud.com/v1/chat/completions — a bare https://api.watt-cloud.com/v1 will not work.

Applying the Changes

Both files are hot-reloaded. Save, wait roughly a second, and the change is live — restarting the tool is unnecessary.

Pick a Model and Start

Open the model dropdown and choose one of the entries you configured — the model is ready to use.