How to Configure OpenClaw: Complete Setup Guide for 2026
OpenClaw (formerly Moltbot/Clawdbot) is the viral open-source AI assistant with 100,000+ GitHub stars. This guide walks you through installation and configuration.
What You'll Need
- Node.js 18+ and pnpm package manager
- An API key from your preferred LLM provider (Anthropic recommended)
- A messaging platform account (WhatsApp, Telegram, Discord, or Slack)
Installation Methods
Method 1: Build from Source (Recommended)
Clone and build the official repository:
bashgit clone https://github.com/openclaw/openclaw.git cd openclaw pnpm install pnpm ui:build # auto-installs UI deps on first run pnpm build
Method 2: DigitalOcean One-Click Deploy
DigitalOcean offers a pre-configured droplet with OpenClaw version 2026.1.24-1 and all dependencies. Perfect for cloud deployment without manual setup.
Method 3: Docker Compose
For containerized deployment:
bashdocker-compose up -d openclaw onboard --install-daemon
Running the Onboarding Wizard
After installation, run the interactive setup:
bashopenclaw onboard --install-daemon
The wizard guides you through:
- Model Provider Setup: Select your LLM (Anthropic, OpenAI, Google, OpenRouter, etc.)
- API Key Entry: Paste your API key or authenticate via OAuth
- Messaging Channels: Connect WhatsApp, Telegram, Discord, or other platforms
- Daemon Configuration: Set up background service for 24/7 operation
Accessing the Control UI
After setup, access the web dashboard:
codehttp://127.0.0.1:18789/?token=YOUR-GENERATED-TOKEN
Replace
with the token displayed at the end of the wizard.codeYOUR-GENERATED-TOKEN
Platform-Specific Notes
macOS
- Install Xcode Command Line Tools if building native apps
- For CLI + gateway only, Node.js is sufficient
Windows
- WSL2 (Ubuntu) is strongly recommended
- Native Windows is untested and has poorer tool compatibility
Linux
- Ubuntu 24.04+ works best
- Ensure you have sudo privileges for daemon installation
Security Best Practices
Running an AI agent with access to your computer raises security concerns. Recommended approaches:
- Dedicated Machine: Run OpenClaw on a separate device, not your main workstation
- Cloud Deployment: Use a cloud server to isolate OpenClaw from personal data
- Credential Isolation: Don't connect production accounts during testing
- Docker Hardening: Use containerization for additional isolation
Cost Considerations
OpenClaw itself is free, but you'll pay for API usage. Expect approximately $30/month for moderate automation tasks using Claude or GPT-4 class models.
Next Steps
Once configured, explore: