OpenClaw is a powerful open-source AI assistant that connects Claude, ChatGPT, Gemini, and local models to your messaging apps — WhatsApp, Telegram, Slack, Discord, and more. It has millions of users and an active GitHub community. But if you have ever tried to install the core OpenClaw project yourself, you know the pain: terminal commands, Node.js version management, WSL2 on Windows, config files, and debugging npm errors.
This guide shows you the easiest way to install and run OpenClaw in 2026 — using OpenClaw Easy, the official desktop app. No terminal. No Node.js. No config files. Under a minute on any machine.
The Two Ways to Install OpenClaw
There are two ways to get OpenClaw running on your machine:
Option A: OpenClaw CLI (core project)
The official OpenClaw repository is a Node.js CLI application. To install it, you need:
- Node.js 22 or later installed on your system
- npm or pnpm for package management
- WSL2 on Windows (the CLI runs best in a Unix-like environment)
- Terminal fluency to run commands and edit config files
- Time — typically 30 to 60 minutes for a first-time setup
This is the right path for developers who want the full CLI experience, custom scripting, or to contribute to OpenClaw's open-source codebase.
Option B: OpenClaw Easy (recommended for everyone else)
OpenClaw Easy is built on top of the same OpenClaw engine, packaged as a native desktop application. You get all of OpenClaw's capabilities through a graphical interface — no terminal, no package manager, no config files.
This guide covers Option B. If you want the full CLI setup, see the official OpenClaw documentation. For everyone else, read on.
Install OpenClaw Easy: Step-by-Step
Choose Your Platform
macOS
Apple Silicon & Intel
Windows
Windows 10 & 11
1 Download OpenClaw Easy
Go to openclaw-easy.com and click the download button for your operating system:
- macOS (Apple Silicon) — downloads a .dmg for M1, M2, M3, and M4 Macs.
- macOS (Intel) — downloads a .dmg for older Intel Macs.
- Windows — downloads a .exe installer for Windows 10 and 11 (64-bit).
Not sure which Mac you have? Click the Apple menu → About This Mac. If it says "Apple M1/M2/M3/M4", download the Apple Silicon version. If it says "Intel", download the Intel version.
2 Install the App
On macOS:
- Open the downloaded .dmg file.
- Drag the OpenClaw Easy icon to the Applications folder.
- Open Applications and double-click OpenClaw Easy.
- If macOS says the app is from an unidentified developer, go to System Settings → Privacy & Security and click Open Anyway.
On Windows:
- Double-click the downloaded .exe file.
- If Windows SmartScreen appears, click More info → Run anyway.
- Follow the installer prompts. Installation completes in about 30 seconds.
- OpenClaw Easy appears in your Start Menu and optionally as a desktop shortcut.
3 Open OpenClaw Easy
Launch the app. You will see the OpenClaw Easy dashboard — a full graphical interface with:
- Channels sidebar — manage WhatsApp, Telegram, Slack, Discord, Feishu, and Line
- Chat interface — test your AI assistant directly from the app
- AI Provider settings — configure which AI model to use
- Agent settings — customize your AI's behavior, system prompt, and tools
- Cron Jobs — schedule automated messages
4 Set Up Your AI Provider
In the sidebar, click AI Provider. You have three options:
- Premium plan (easiest) — subscribe in-app and get managed AI access. No API keys to handle, billing is simple, and the app takes care of everything.
- Bring Your Own Key (BYOK) — paste an API key from your existing account at Anthropic (Claude), OpenAI (ChatGPT), Google (Gemini), or DeepSeek. This option is completely free — you only pay your AI provider's standard per-token rates.
- Local LLM via Ollama — run AI entirely offline on your own machine. Install Ollama, pull a model like Llama 3 or Qwen 2.5, and point OpenClaw Easy at it. No internet required for inference and no data leaves your machine.
Recommendation for beginners: Start with BYOK using an OpenAI or Anthropic API key. Both are easy to create, and costs are typically a few cents per day for normal personal use.
5 Connect Your First Messaging Channel
Go to Channels in the sidebar. Here is how to connect the most popular options:
WhatsApp (easiest)
- Click WhatsApp in the channels list.
- A QR code appears on screen.
- On your phone, open WhatsApp → Settings → Linked Devices → Link a Device.
- Scan the QR code.
- Done — your AI assistant is live on WhatsApp.
Telegram
- Open Telegram and message @BotFather.
- Send /newbot and follow the prompts to create a bot.
- Copy the bot token BotFather gives you.
- In OpenClaw Easy, click Telegram, paste the token, and click Connect.
Slack / Discord
Both require creating an app in the respective developer portals. OpenClaw Easy guides you through the steps in-app. See the dedicated guides for Slack and Discord.
That Is It — OpenClaw Is Running
You have installed OpenClaw without touching a terminal. Your AI assistant is now live on your chosen messaging channel, running entirely on your computer. Every message your contacts send gets answered by the AI model you chose.
OpenClaw Easy runs in the system tray (macOS menu bar / Windows taskbar) so it stays active in the background without keeping a window open.
What Can You Do Next?
Connect More Channels
You can connect multiple channels simultaneously. Run your AI assistant on WhatsApp and Telegram at the same time, with the same AI provider, from the same app.
Customize the AI Behavior
Go to Agent Settings to write a system prompt that defines your AI's personality and expertise. For example:
- "You are a helpful personal assistant. Keep responses concise and friendly."
- "You are a customer support agent for my small business. Answer questions about our products based on the context I provide."
- "You are a coding assistant. Format code in code blocks and explain your reasoning."
Schedule Automated Messages
Use Cron Jobs to set up scheduled AI tasks. Send a daily morning briefing, a weekly summary, or automated reminders on any channel. See the cron jobs guide for details.
Switch to a Local LLM
If privacy is a priority, switch to a local model. Install Ollama, pull a model, and your AI runs entirely on your machine — no API calls, no data leaving your computer. See the local LLM guide.
Frequently Asked Questions
Is OpenClaw Easy the same as OpenClaw?
OpenClaw Easy is built on top of OpenClaw — the same core engine. The difference is the interface: the core OpenClaw project is a CLI tool, while OpenClaw Easy wraps it in a desktop GUI. You get the same features, same channels, and same AI capabilities without needing a terminal.
Is it free?
Yes. OpenClaw Easy is free to download. Use it with your own API keys at no charge. Premium plans are available in-app for users who prefer managed AI access.
Does it work offline?
The app itself works offline. If you use a cloud AI provider (Claude, ChatGPT, Gemini), those API calls require internet. If you use Ollama with a local model, inference is completely offline.
What happened to Clawdbot and Moltbot?
OpenClaw was previously known as Clawdbot and before that Moltbot. The project was rebranded to OpenClaw. OpenClaw Easy works with the current version. If you were searching for a Clawdbot or Moltbot desktop app, OpenClaw Easy is what you are looking for.
Does it support the latest Claude / GPT-4o models?
Yes. OpenClaw Easy updates alongside the core OpenClaw project and supports the latest models from Anthropic, OpenAI, Google, and other providers as they are added to the platform.
What Is Next?
- Windows users — read the dedicated OpenClaw for Windows guide for Windows-specific setup tips.
- WhatsApp — full guide to adding AI to WhatsApp with screenshots.
- Telegram — create a Telegram AI bot without a server in 5 minutes.
- Local AI — run OpenClaw with Ollama for full privacy.
- Privacy — learn why local-first AI is the right choice for personal data.
OpenClaw is one of the most capable AI assistant platforms available today, and it is free and open source. OpenClaw Easy makes it accessible to everyone — not just developers. Download it free and have OpenClaw running in under a minute.