Skip to main content

OpenClaw Usage

OpenClaw is an open-source AI agent project developed by Peter Steinberger. It can be deployed on any device and any OS as a personal AI assistant.

Install OpenClaw

Install OpenClaw using the installation script.

Device
curl -fsSL https://openclaw.ai/install.sh | bash

Configure OpenClaw

tip

If you install with the script, you will be guided through the configuration during installation. If you want to reconfigure, run the following command.

Device
openclaw onboard --install-daemon
  • Risk disclaimer

    Select Yes.

  • Onboarding mode

    Select QuickStart.

  • Model provider

    Choose a provider based on the model API key you already have.

    If you don't have an API key, you can use Qwen OAuth (free). Follow the link to log in/register a qwen.ai account to complete OAuth authorization.

    After authorization, select the default qwen-portal/coder-model model.

  • Configure channels

    Match OpenClaw with the social apps you commonly use. Follow the on-screen prompts to configure; details are omitted here.

  • Configure skills

    Select yes.

  • Choose how to install Homebrew

    Select npm.

  • Install missing skill dependencies

    Choose according to your needs.

  • Configure API keys used by each skill

  • Start the interactive UI

    In How do you want to hatch your bot?, you can choose between two interaction modes: TUI (terminal UI) or Web UI.

    • Web UI (recommended)

      tip

      Port: 18789

    • TUI (terminal)

More features

For more usage details, see the official OpenClaw documentation.

    You need to be logged into GitHub to post a comment. If you are already logged in, please ignore this message.

    Radxa-docs © 2026 by Radxa Computer (Shenzhen) Co.,Ltd. is licensed under CC BY 4.0