Start Here

This section is the shortest path into OpenClaw.

OpenClaw gives you a ready-to-use scientific runtime, a shared workspace for people and agents, and a calmer way to move from setup to outputs.

You do not need to understand Docker deeply to begin

Start by treating the container as a portable lab bench: it holds the tools. The repository is the lab notebook: it holds the memory.

The Core Mental Model

flowchart LR
  GitHub["GitHub\ncontrol plane"] --> Repo["Repository\nproject memory"]
  Repo --> Container["Container\nruntime and tools"]
  Secrets["Secrets\ninjected locally"] --> Container
  Storage["External storage\nlarge data and outputs"] --> Container
  Container --> Workspace["Workspace\nactive scientific work"]
  Workspace --> Review["Human review\nbefore publishing"]

Repeat this when you feel lost:

GitHub = control plane. Repo = memory. Container = runtime. Secrets are injected, never stored. External storage holds large durable data.

First Path

1. Read What is OpenClaw?.
2. Walk through First 10 Minutes.
3. Launch the workspace with Launch Locally.
4. Learn Where Files Go.
5. Use Troubleshooting if the browser, token, or startup flow feels strange.

The calm command loop is:

make init-working-group
make doctor
make checkpoint

User Modes

Mode	Start With	Main Concern
Everyday Scientist	What is OpenClaw?	Understanding the system without infrastructure overload
Working Group Lead	Template Mode	Making a reusable working group from the base image
Data/Workflow Maintainer	Where Files Go	Data placement, provenance, and reproducible outputs
Infrastructure Admin	Operations	Ports, credentials, startup, and deployment
Developer/Customizer	Architecture	Extending agents, docs, branding, and workflows