Configuration

Nilux AI reads settings from a configuration file and environment variables. This page documents all available options.

Configuration File

The main configuration file is located at:

~/.nilux/config.json

If the file does not exist, Nilux AI creates it with defaults on first run.

Available Options

Example Configuration

{
  "api_key": "sk_live_xxxxxxxxxxxxxxxxxxxxxxxx",
  "model": "standard",
  "theme": "dark"
}

Environment Variables

You can also configure Nilux AI via environment variables. Environment variables take precedence over the config file.

Using Environment Variables

export NILUX_API_KEY="sk_live_xxxxxxxxxxxxxxxxxxxxxxxx"
export NILUX_MODEL="pro"
nilux

Tip: Use environment variables in CI/CD pipelines or when you need to temporarily override settings without editing the config file.

Model Selection

You can switch models at any time using the /model slash command within the CLI:

/model pro

Available models are documented on the Models page.

API Key

Your API key authenticates you to the Nilux backend. You can find or rotate your key from the API Keys dashboard page.

Warning: Never commit your config.json or API key to version control. The ~/.nilux/ directory is outside your project tree by design.

Project-Specific Settings

For project-specific overrides, you can create a .nilux/ directory in your project root. Settings here merge with your global config, with project values taking priority.

your-project/
  .nilux/
    skills/         # Project-specific skills
    agents/         # Project-specific custom agents

This is covered in detail in the Skills and Agents documentation.

Next Steps

• Quickstart — Make your first request
• Models — Choose the right model for your task
• API Keys — Manage your API key