Configuration

The CLI stores configuration in ~/.promptodex/config.json.

Interactive setup

The easiest way to configure:

pod config

Manual configuration

Create or edit ~/.promptodex/config.json:

{
  "apiKey": "your-promptodex-api-key",
  "defaultModel": "4.1",
  "vendors": {
    "openai": {
      "apiKey": "sk-your-openai-key"
    },
    "anthropic": {
      "apiKey": "sk-your-anthropic-key"
    },
    "xai": {
      "apiKey": "xai-your-key"
    },
    "localhost": {
      "port": 11434
    }
  },
  "models": {
    "4.1": {
      "vendor": "openai",
      "model": "gpt-4.1"
    },
    "sonnet": {
      "vendor": "anthropic",
      "model": "claude-sonnet-4"
    },
    "grok": {
      "vendor": "xai",
      "model": "grok-3"
    },
    "llama": {
      "vendor": "localhost",
      "model": "llama3.2"
    }
  }
}

Config fields

Field	Description
apiKey	(Optional) Your Promptodex API key for accessing private prompts
defaultModel	The model alias used when no model is specified
vendors	API keys/config for each AI provider
models	Model aliases mapping to a vendor and model ID

Supported vendors

Vendor	Config	Models
OpenAI	apiKey	GPT-4.1, GPT-4o, o1, etc.
Anthropic	apiKey	Claude Sonnet 4, Claude Opus 4, etc.
xAI	apiKey	Grok-3, Grok-2, etc.
Localhost	port	Ollama, LMStudio, any OpenAI-compatible server

Model resolution

The CLI picks a model in this order:

1. --model flag (if specified)
2. Prompt's recommended model (if the prompt suggests one)
3. defaultModel from config

Model aliases

You define model aliases in the models section. Each alias maps to a vendor and model ID:

{
  "models": {
    "fast": {
      "vendor": "openai",
      "model": "gpt-4.1-mini"
    }
  }
}

Then use it:

pod summarize --model fast

View current config

pod show-config

Troubleshoot config

pod doctor

This checks if your config is valid and your API keys are working.