OpenClaw — Personal AI Assistant
OpenClaw — The AI that actually does things. Your personal assistant on any platform.
DevOps Tutorial
Summary
This guide details the process of deploying OpenClaw (Clawdbot) within a Docker container to avoid running it directly on a host machine. It covers cloning the repository, executing the setup script to create volume-mounted directories, and configuring the initial onboarding wizard.
Highlights
- OpenClaw supports Docker out of the box via a provided docker-setup.sh script and docker-compose.yml.
- The setup creates two local directories (~/.openclaw and ~/openclaw/workspace) mounted as volumes for configuration and agent workspace.
- Onboarding requires selecting 'manual' mode and a local gateway, with specific authentication steps for OpenAI ChatGPT OAuth.
- Administrative tasks can be performed using the separate 'openclaw-cli' container defined in the Docker Compose setup.
- The web UI runs on port 18789 and requires authentication via a token URL parameter.
auto-generated
Context
Audience
Software Developers
DomainSoftware Engineering
Formattechnical blog post
Accessfree online
Source Overview
OpenClaw can be run in Docker by cloning the GitHub repository and executing the docker-setup.sh script, which creates configuration and workspace directories mounted as volumes.
The setup wizard prompts for onboarding mode, model provider selection, and optional integrations like Tailscale. The web UI runs on port 18789 and requires authentication via a token URL parameter, while administrative commands can be executed through the openclaw-cli container.
Topics
Related
OpenClawDockerDocker ComposeSimon WillisonLLM Deployment
