To install Clawdbot on your system, you first need a supported runtime (Node.js ≥ 22) and a shell environment where you can run command-line tools. Clawdbot is distributed as a global CLI package that you install via npm or through an installer script, and it works on macOS, Linux, or Windows via WSL2. The simplest method is to use the installer script provided by the Clawdbot project; this script detects your operating system, ensures the correct Node version is available, installs the CLI, and launches the initial onboarding setup. Once installed, the clawdbot command becomes available globally, letting you invoke onboarding, start the Gateway daemon, and interact with agents directly from your terminal.
Here is a concrete example of installing Clawdbot on a typical Unix-like system. Open your terminal and run the installer script:curl -fsSL https://clawd.bot/install.sh | bashThis will download and execute the installer, set up the CLI tools, and usually run the onboarding wizard afterward. If you already have Node.js installed and prefer to skip the script, you can install the package directly with npm:npm install -g clawdbot@latest If you encounter build issues with native dependencies such as sharp, you might need to set an environment variable like SHARP_IGNORE_GLOBAL_LIBVIPS=1 before installation, or install build tools such as node-gyp. After installation, confirm that clawdbot is in your PATH by running clawdbot --version.
On Windows, the recommended approach is to install and use WSL2 with a Linux distribution (Ubuntu is common) and then run the commands above inside that environment. The Clawdbot CLI and Gateway are consistently maintained in a Linux-like environment, which avoids many Windows-specific issues. Once installed, you can manage Clawdbot using commands like clawdbot onboard for setup and clawdbot gateway to run the service. Keep in mind that Clawdbot itself doesn’t bundle an AI model; you’ll need to configure access to a model provider or backend of your choice during setup, and you can later enhance Clawdbot with integrations such as storing and querying long-term context in a vector database like Milvus or using a managed instance on Zilliz Cloud if your workflows require semantic retrieval.