How to Install OpenClaw on Windows

Install OpenClaw on Windows using PowerShell or WSL2. This guide explains how each method works, so you can choose the right setup for your needs. If you prefer a simpler option, you can also use Kimi Claw to run your agent in the cloud without manual setup.Try Kimi Claw
14 min read·2026-03-23

OpenClaw is a powerful tool for building AI workflows, but setting it up on Windows can be confusing for beginners. This guide walks you through the main installation methods so you can get started quickly.

Table of contents

OpenClaw Windows installation: 3 methods compared

Before diving into the steps, here is a quick overview of the three ways to install OpenClaw on Windows. Each method has different trade-offs in terms of setup time, feature coverage, and technical requirements.

PowerShellWSL2Kimi Claw (Online)
How it worksNative Windows installer via command lineFull Linux environment inside WindowsAn online tool that lets users run OpenClaw without local setup
Setup complexityLow to medium, requires running commands and following setup stepsHigh, requires Linux setup, system configuration, and dependenciesVery low, no setup required
Code or terminal requiredYesYesNo
Always-on availabilityOnly while your PC is runningOnly while WSL is runningAvailable 24/7, runs on cloud servers
Best forDevelopers comfortable with CLIDevelopers needing full Linux environmentNon-technical users & teams

Kimi Claw is an online tool that lets you run OpenClaw in your browser. It handles setup and configuration automatically, so you do not need Node.js, WSL, or API key setup. You can start using a ready-to-use agent within minutes.

For a local PowerShell or WSL2 setup, scroll down to Method 1 or Method 2. To skip local installation entirely, jump to Method 3: Kimi Claw. If you want to get started faster, you can begin right away.

What you need before installing OpenClaw on Windows locally

If you plan to install OpenClaw locally, make sure your system meets the following requirements. If you choose the cloud option like Kimi Claw, you can skip this section entirely.

System requirements

  • Windows 10 or Windows 11 (64-bit)
  • At least 4 GB RAM (8 GB recommended for stable npm builds)
  • Stable internet connection
  • Administrator privileges for initial setup if you choose WSL2
  • For OpenClaw WSL2: virtualization must be enabled in BIOS/UEFI

Required software

  • Node.js 22 or higher. OpenClaw requires a modern Node.js runtime. Download it from nodejs.org or install via the onboarding script.
  • Git for Windows (optional but recommended). Some npm packages and skills depend on Git.

Method 1: Install OpenClaw on Windows using PowerShell

Compared to WSL2, the PowerShell method requires fewer steps and no Linux environment. A single command downloads OpenClaw, checks your Node.js version, and launches the onboarding wizard. This method works on both Windows 10 and Windows 11.

Step 1: Install and run your OpenClaw

Windows restricts script execution by default to protect your system. Before installing OpenClaw, you must allow PowerShell to execute local scripts by running this "safety switch" command:

Set-ExecutionPolicy RemoteSigned -Scope CurrentUser

Then, run the following command:

iwr -useb https://openclaw.ai/install.ps1 | iex

After running the command, you should see output similar to the screenshot below. Close and reopen PowerShell to refresh your system PATH before proceeding to onboarding.

Step 2: Ensure risky note & select onboarding mode

Select Yes to acknowledge the risk and continue.

Next, select QuickStart to quickly start onboarding. You can adjust settings later, or choose Manual for full configuration.

Step 3: Configure and authenticate your AI model

  1. Select Moonshot AI as model/auth provider

  1. Authenticate and configure your AI model

To connect OpenClaw, activate your Kimi API Key via the Kimi Platform. While a $5 recharge earns you a $5 bonus voucher, we recommend $20+ to unlock Tier 2 access. This reduces latency and ensures that your OpenClaw agent handles complex, multi-step tasks.

  1. Create a Kimi API key

Access the API Keys section and click Create API Key. Note that you should keep your API Key in a secure and accessible place. It only shows once when you create it.

  1. Enter your API key and select the default model

When the onboarding wizard prompts you, paste your unique API key into the PowerShell/Terminal window.

When asked to choose a default model, select moonshot/kimi-k2.5 (or choose "Keep current" if it's already highlighted). This model provides the best balance of speed and reasoning for agentic tasks.

Step 4: Select channel

Choose which app you want to connect to your OpenClaw. If you have no idea, just skip it.

Step 5: Configure for Performance

Step 6: Enable Hooks

Choose a method to chat with your bot, then wait for it to start running.

Step 7: Run Your OpenClaw

After all the above is done, enjoy your OpenClaw.

OpenClaw Web UI running on Windows after successful installation

Method 2: Install OpenClaw on Windows with WSL2

OpenClaw WSL2 setup provides a full Linux environment inside Windows, which is the officially recommended way to run OpenClaw. This method gives you access to the complete feature set, including systemd daemon management, native shell scripting, and smoother npm builds. If you plan to use OpenClaw on Windows for daily workflows or development, OpenClaw WSL is the best choice.

Step 1: Enable WSL2 for OpenClaw on Windows 11/10

Open PowerShell as Administrator and run:

wsl --install -d Ubuntu

This command enables WSL, downloads Ubuntu, and installs it. After the process completes, restart your computer. When Windows reboots, Ubuntu will launch and ask you to create a username and password.

If you already have WSL installed, verify it is running version 2:

wsl --list --verbose

Step 2: Configure WSL2 for systemd

OpenClaw runs as a background service using systemd. Enable it in your WSL configuration:

sudo tee /etc/wsl.conf > /dev/null << 'EOF' [boot] systemd=true [interop] enabled=true appendWindowsPath=true EOF

Then restart WSL from PowerShell:

wsl --shutdown

Reopen your Ubuntu terminal and verify systemd is running:

ps -p 1 -o comm=

The output should show systemd. If it shows init instead, double-check your /etc/wsl.conf file.

Step 3: Install Node.js 22+

OpenClaw requires Node.js 22 or later. Install it from the NodeSource repository:

curl -fsSL https://deb.nodesource.com/setup_22.x | sudo -E bash - sudo apt-get install -y nodejs

Verify the installation:

node --version # Should show v22.x.x or higher npm --version

Step 4: Install OpenClaw via npm

With Node.js ready, install OpenClaw globally:

sudo npm install -g openclaw@latest

Alternatively, use the official installer script:

curl -fsSL https://openclaw.ai/install.sh | bash

Verify the installation:

openclaw --version

Step 5: Complete OpenClaw Windows setup via onboarding

Run the onboarding wizard using the daemon flag to set up OpenClaw as a systemd service:

openclaw onboard --install-daemon

The wizard will guide you through:

  • Selecting your LLM provider and entering your API key
  • Choosing a default model
  • Configuring messaging channels (optional)
  • Installing the gateway as a background service

After onboarding, enable service persistence, so OpenClaw starts automatically when WSL boots:

sudo loginctl enable-linger $USER

The gateway is the core background service that handles all OpenClaw operations, including agent requests, messaging, and model execution. Ensuring it is running confirms that your OpenClaw installation is complete and ready for use, and prevents errors in subsequent steps.

Run the following command in your PowerShell or WSL terminal to check its status:

openclaw gateway status

Method 3: Run OpenClaw on Windows with Kimi Claw (no setup required)

If managing Node.js, WSL, and API keys feels like more setup than you need, Kimi Claw offers a simpler way to run OpenClaw. It lets you run OpenClaw directly in your browser without local setup or manual configuration. To get started, follow these steps:

Step 1: Open Kimi Claw and create an instance

Go to the Kimi Claw page and click Create to start.

Create a new OpenClaw instance on Kimi Claw cloud platform

Step 2: Start the deployment

Confirm the deployment in the pop-up. Kimi Claw sets up your OpenClaw environment in about two minutes, including the gateway, workspace, and tools. No commands or manual configuration are required.

Kimi Claw deploying the OpenClaw environment automatically in the cloud

Step 3: Work in your workspace

Once deployment finishes, your workspace opens in the dashboard. You can start chatting with your AI agent, set up skills, configure scheduled tasks, and manage files all from the browser.

Access the Kimi Claw workspace

What you get with Kimi Claw

  • Zero configuration: No Node.js, WSL, or API key management required
  • 40 GB cloud storage: Save reports, datasets, and generated files that persist across sessions
  • Persistent memory: Your AI agent remembers past interactions, preferences, and context
  • 5,000+ ClawHub skills: Access automation, research, coding, and analysis skills instantly
  • Scheduled task automation: Set up recurring tasks like daily reports, reminders, and monitoring
  • Works from any device: Access your OpenClaw agent from any browser on Windows, Mac, or mobile

Windows firewall and security settings for OpenClaw

Kimi Claw automatically handles security and networking on cloud servers. If you prefer to install OpenClaw locally with PowerShell or WSL2, there are a few recommended security settings to configure after installation.

  • Configuring Windows Firewall

When the OpenClaw gateway starts for the first time, Windows Firewall may prompt you to allow network access. Allow it for private networks only. If you are using OpenClaw WSL2, the gateway runs inside the Linux environment and communicates through localhost, so additional firewall rules are usually not needed.

If Browser Relay or other features cannot connect, make sure that localhost traffic is not being blocked by third-party security software such as antivirus programs, VPN clients, or corporate firewalls.

  • Securing the Control UI

The OpenClaw Control UI runs on port 18789 and provides full access to your agent. Never expose this port to the public internet. Keep the gateway bound to 127.0.0.1 and access it only from your local machine.

For secure remote access, use Tailscale to create a private network tunnel:

{ "gateway": { "tailscale": { "mode": "serve" } } }

This keeps the port on localhost while making it accessible through your private Tailscale network.

Common OpenClaw installation errors on Windows and how to fix them

Even with a clear guide, you may run into issues during installation. Below are the most frequently reported problems when installing OpenClaw on Windows and how to fix them.

OpenClaw is not recognized

  • Cause: The npm global bin directory is not in your system PATH.
  • Fix: Run the following command to find your npm prefix
npm prefix -g

Add the output path with \bin appended to your Windows environment variables. Then close and reopen PowerShell.

Git not found during installation

  • Cause: Git for Windows is not installed.
  • Fix: Download and install Git from git-scm.com. After installation, restart PowerShell and retry the OpenClaw Windows installer.

Node.js version is too old

  • Cause: Your system has an older version of Node.js that does not meet the minimum requirement (v22+).
  • Fix: Download the latest LTS version from nodejs.org and install it. If you are using OpenClaw WSL2, reinstall via the NodeSource repository.

Permission denied errors

  • Cause: PowerShell does not have sufficient permissions, or the npm global directory is not writable.
  • Fix: Run PowerShell as Administrator, or change the npm global prefix to a user-writable directory
npm config set prefix "$env:APPDATA\npm"

Gateway won't start

  • Cause: Port 18789 is already in use, or the configuration file is corrupted.
  • Fix: Check for conflicting processes and restart
# Check what is using the port netstat -ano | findstr :18789 # Restart the gateway openclaw gateway restart

WSL2 systemd not working

  • Cause: The /etc/wsl.conf file is missing the systemd configuration.
  • Fix: Verify the file contains the correct settings
cat /etc/wsl.conf # Should include: # [boot] # systemd=true

If the setting is missing, add it and restart WSL with wsl --shutdown in PowerShell.

Conclusion

Installing OpenClaw on Windows can range from a simple script-based setup to a more advanced WSL2 environment, depending on your needs. If you want full control and flexibility, PowerShell and WSL2 are solid options. If you prefer to skip setup and start immediately, Kimi Claw lets you run OpenClaw in your browser without local setup. Choose the approach that fits your workflow and technical comfort level, and you will be ready to get the most out of OpenClaw.

Questions & Answers

Can I install OpenClaw on Windows without WSL?
Yes. OpenClaw provides a native PowerShell installer that runs directly on Windows. Open PowerShell and run iwr -useb https://openclaw.ai/install.ps1 | iex to install. This method requires fewer steps, but some features, like systemd service management, work better in a Linux environment. OpenClaw WSL2 remains the recommended local option for daily use. Kimi Claw is another option that requires no local installation.
Why is WSL2 recommended over PowerShell when running OpenClaw on Windows?
Running OpenClaw in WSL2 provides a full Linux environment, where features such as background services (via systemd), shell scripting, and npm package builds behave more consistently. The PowerShell installer is quicker to set up but runs in a more limited environment with less Linux compatibility. If you want the full OpenClaw experience on Windows without dealing with these setup differences, Kimi Claw lets you run OpenClaw in your browser without local setup.
How do I fix "openclaw is not recognized" on Windows?
This error means the npm global bin directory is not in your system PATH. Run npm config get prefix in PowerShell to find the npm prefix, then add that path to your Windows environment variables. Restart PowerShell after making the change. This is one of the most commonly reported issues with OpenClaw Windows installations and affects only local installs.
Does OpenClaw keep running when my Windows PC is off or asleep?
No. If you install OpenClaw locally, the agent stops when your computer shuts down or enters sleep mode. WSL2 also suspends when Windows sleeps, which means scheduled tasks and background services will not run during that time. To keep OpenClaw available 24/7 without leaving your machine on, Kimi Claw runs your agent on cloud servers that stay online continuously.
Does OpenClaw work on Windows?
Yes. OpenClaw runs on Windows through a native PowerShell installer or WSL2. OpenClaw WSL2 is the recommended local approach because it supports the full feature set, including systemd background services and native Linux shell tools. Both Windows 10 and Windows 11 are supported. For users who want to skip local setup, Kimi Claw runs OpenClaw on cloud servers and is accessible from any browser.