Running OpenClaw Using Docker Desktop: Your Personal AI Assistant in a Container

Your own personal AI assistant, running entirely on your machine, talking to you through WhatsApp, Telegram, Slack, or Discord. Does it sound like science fiction?

Meet OpenClaw, the open-source, self-hosted AI agent framework that's taking the developer world by storm. And the best part? You can get it up and running in minutes using Docker Desktop.

In this guide, I'll walk you through everything you need to know about running OpenClaw with Docker Desktop - from understanding what OpenClaw is, to setting it up, configuring messaging channels, managing it day-to-day, and troubleshooting common issues.

What is OpenClaw?

OpenClaw (formerly known as MoltBot/Clawdbot) is a free, open-source personal AI assistant framework created by Peter Steinberger. It's designed to run on your own devices and answer you on the channels you already use — WhatsApp, Telegram, Slack, Discord, Signal, iMessage, Microsoft Teams, and more.

Think of it as your always-on AI teammate that can:

• Read files, run commands, and automate tasks across your workspace
• Maintain persistent memory across conversations
• Connect to 50+ integrations including GitHub, Gmail, Spotify, Obsidian, and more
• Run scheduled jobs via cron for proactive automation
• Support multiple AI providers — Anthropic Claude, OpenAI GPT, Google Gemini, Groq, Mistral, and others
• Execute browser automation via CDP-controlled Chrome/Chromium
• Render a live Canvas for agent-driven visual workspaces

The project has exploded in popularity with over 37,000 stars on GitHub, and its community is growing rapidly.

Why Docker?

Running OpenClaw in Docker gives you several advantages:

1. Isolation — OpenClaw can read files, run commands, and automate tasks. Docker keeps all of this contained in a sandboxed environment, preventing accidental changes to your host system.
2. Reproducibility — The exact same environment every time, regardless of your host OS.
3. Easy cleanup — Don't like it? docker compose down and it's gone.
4. No dependency conflicts — No need to install Node.js, npm, or any other tools on your host machine.
5. Security — The default Docker image runs as a non-root node user, minimizing the attack surface.

Prerequisites

Before we begin, make sure you have:

• Docker Desktop installed and running (macOS, Windows, or Linux)
  • Download from docker.com/products/docker-desktop
• Git installed on your system
• Bash v4+ (macOS ships with an older version — upgrade via brew install bash if needed)
• At least 2 GB RAM (4 GB recommended)
• 10 GB+ disk space for Docker images and application data
• An API key from at least one AI provider (Anthropic, OpenAI, Google, etc.)

Verify Docker Installation

Open your terminal and run:

docker --version
# Docker version 27.x.x, build xxxxxxx

docker compose version
# Docker Compose version v2.x.x

If both commands return version numbers, you're good to go.

Step 1: Clone the OpenClaw Repository

Start by cloning the official OpenClaw repository:

git clone https://github.com/openclaw/openclaw.git
cd openclaw

This downloads the entire OpenClaw codebase, including the Dockerfile, docker-compose.yml, and the docker-setup.sh script.

Step 2: Run the Docker Setup Script

OpenClaw provides an all-in-one setup script that handles everything:

./docker-setup.sh

What This Script Does

Behind the scenes, docker-setup.sh performs these steps:

1. Builds the gateway image locally from the included Dockerfile
2. Runs the onboarding wizard interactively inside a Docker container
3. Generates a gateway token for accessing the Control UI
4. Creates configuration directories on your host
5. Starts the gateway via Docker Compose
6. Writes the token to .env for future reference

Note: Unlike many Docker applications where you pull a pre-built image, OpenClaw builds the image locally on your server. This means you're always running the exact version from the repository.

The Onboarding Wizard

During the setup, you'll be guided through an interactive wizard. Here's what to expect:

==> Onboarding (interactive)
When prompted:
  - Gateway bind: lan
  - Gateway auth: token
  - Gateway token: <auto-generated>
  - Tailscale exposure: Off
  - Install Gateway daemon: No

The wizard will ask you to:

1. Choose your AI provider — Enter your API key for Anthropic, OpenAI, Google Gemini, or another supported provider
2. Set gateway binding — Choose lan for local network access
3. Configure authentication — Token-based auth is recommended
4. Set up messaging channels (optional at this stage)

Configuration Directories

After setup, OpenClaw creates two important directories on your host:

These directories are mounted as volumes inside the Docker container, so your data persists across container restarts.

Step 3: Access the Control UI

Once the gateway is running, open your browser and navigate to:

http://127.0.0.1:18789/

You'll need the gateway token to authenticate. If you missed it during setup, retrieve it with:

docker compose run --rm openclaw-cli dashboard --no-open

This will output a URL with the token parameter — something like:

http://127.0.0.1:18789/?token=your-secret-token-here

If you see "unauthorized" or "disconnected (1008): pairing required", follow these steps:

# Get a fresh dashboard link
docker compose run --rm openclaw-cli dashboard --no-open

# List pending device requests
docker compose run --rm openclaw-cli devices list

# Approve your browser device
docker compose run --rm openclaw-cli devices approve <requestId>

Step 4: Connect a Messaging Channel

OpenClaw truly shines when you connect it to a messaging platform. Let's set up Telegram as an example — it's the quickest to configure.

Setting Up Telegram

1. Create a Telegram Bot

Open Telegram and chat with @BotFather:

/newbot

Follow the prompts to name your bot. BotFather will give you a bot token — save it.

2. Add the Channel to OpenClaw

docker compose run --rm openclaw-cli channels add \
  --channel telegram \
  --token <YOUR_BOT_TOKEN>

3. Pair Your Account

Send any message to your new bot on Telegram. OpenClaw will respond with a pairing code. Approve it:

docker compose run --rm openclaw-cli pairing approve telegram <CODE>

That's it! You can now message your AI assistant directly from Telegram on your phone.

Other Channel Options

# WhatsApp (QR code based)
docker compose run --rm openclaw-cli channels login

# Discord
docker compose run --rm openclaw-cli channels add \
  --channel discord \
  --token <DISCORD_BOT_TOKEN>

# Slack
docker compose run --rm openclaw-cli channels add \
  --channel slack \
  --token <SLACK_BOT_TOKEN>

Step 5: Verify Everything is Running

Check Container Status

docker compose ps

You should see output similar to:

NAME                          IMAGE            STATUS
openclaw-openclaw-gateway-1   openclaw:local   Up 2 minutes

Health Check

docker compose exec openclaw-gateway \
  node dist/index.js gateway health

View Logs

# Follow logs in real-time
docker compose logs -f openclaw-gateway

# Last 100 lines only
docker compose logs --tail 100 openclaw-gateway

Understanding the Architecture

Here's how OpenClaw runs inside Docker:

WhatsApp / Telegram / Slack / Discord / Signal / iMessage / Teams
                          │
                          ▼
              ┌───────────────────────┐
              │       Gateway         │
              │   (control plane)     │
              │  ws://127.0.0.1:18789 │
              └───────────┬───────────┘
                          │
                ├─ Pi agent (RPC)
                ├─ CLI (openclaw …)
                ├─ WebChat UI
                ├─ Canvas + A2UI
                └─ Browser (CDP)

The Gateway is the single control plane that manages sessions, channels, tools, and events. It communicates with your chosen AI provider (Claude, GPT, etc.) and routes messages between you and the agent through your preferred messaging channels.

Docker Compose Services

The docker-compose.yml defines two main services:

Configuring AI Providers

OpenClaw is model-agnostic. You can configure multiple providers and switch between them. Here are the supported providers:

You can provide API keys during the onboarding wizard, or set them as environment variables in your Docker Compose configuration.

Security Best Practices

OpenClaw is powerful — it can read files, execute commands, and automate tasks. Here are important security considerations:

1. Docker Provides Isolation

Running in Docker means the agent cannot access your host filesystem outside of the mounted volumes. This is a significant safety advantage over running directly on your machine.

2. Default Network Is "None"

By default, OpenClaw's Docker configuration uses network: none, meaning no egress traffic. You must explicitly enable networking if the agent needs internet access.

3. Non-Root Execution

The default image runs as the node user (UID 1000), not root. This limits what the agent can do even inside the container.

4. Run the Doctor Command

OpenClaw includes a built-in security audit:

docker compose exec openclaw-gateway \
  node dist/index.js doctor

This surfaces risky or misconfigured DM policies.

5. Per-Agent Sandboxing

For multi-agent setups, you can configure per-agent sandbox profiles with isolated workspaces and per-agent sessions:

# Enable sandboxing
docker compose exec openclaw-gateway \
  node dist/index.js config set sandbox.enabled true

Docker Desktop Tips

Resource Allocation

For optimal OpenClaw performance in Docker Desktop:

• Memory: Allocate at least 4 GB (Settings → Resources → Memory)
• CPUs: 2+ cores recommended
• Disk: Ensure at least 10 GB available for images

File Sharing

If you're mounting host directories, verify they're shared:

• macOS/Windows: Docker Desktop → Settings → Resources → File Sharing
• Add any paths you're using in OPENCLAW_EXTRA_MOUNTS

Using Docker Desktop Dashboard

Docker Desktop provides a visual interface to manage your OpenClaw containers:

1. Open Docker Desktop
2. Click on Containers in the left sidebar
3. You'll see openclaw-openclaw-gateway-1 running
4. Click on it to view logs, inspect environment variables, and monitor resource usage
5. Use the Terminal tab to open a shell inside the container

What Can You Do With OpenClaw?

Once your assistant is running, the possibilities are vast. Here are some real-world examples:

• Personal knowledge management — "Summarize the research paper I uploaded and add it to my notes"
• Code review and development — "Review this PR and suggest improvements"
• Automated research — "Search for the latest developments in container security and compile a report"
• Task scheduling — "Every Monday at 9 AM, check my calendar and send me a summary on Telegram"
• Content pipelines — "Draft a blog post about Docker best practices based on our previous discussions"
• Financial tracking — "Parse my bank statement and categorize expenses"

The community has built hundreds of skills that extend OpenClaw's capabilities even further.

Wrapping Up

OpenClaw running on Docker Desktop gives you the best of both worlds: the power of a personal AI assistant with the safety and isolation of containerization. Whether you're a developer looking to automate workflows, a content creator wanting an AI collaborator, or just someone curious about running their own AI agent, Docker makes the experience smooth and secure.

The setup takes about 15–30 minutes, and once it's running, you've got an always-on AI assistant that lives on your machine, respects your privacy, and talks to you wherever you are.

Useful Links:

• OpenClaw GitHub Repository
• Official Docker Documentation
• OpenClaw Website
• Docker Desktop Download

Give it a try and join the lobster revolution! 🦞