{/* Last updated: 2026-02-10 | OpenClaw v latest | Node.js 22+ | MIT License */}
Version note: This guide covers OpenClaw (formerly Clawdbot, then Moltbot) as of February 2026. The project has 185,000+ GitHub stars and is one of the fastest-growing open-source projects in history. OpenClaw is under active development — features and commands may change. Always check docs.openclaw.ai for the latest documentation.
What Is OpenClaw and Why It Matters
OpenClaw is a free, open-source, self-hosted personal AI assistant that runs on your own devices. Unlike cloud-based assistants like ChatGPT or Claude's web interface, OpenClaw keeps your data local and connects to the messaging platforms you already use — WhatsApp, Telegram, Slack, Discord, and 9 more channels.
But OpenClaw isn't just a chatbot. It's an AI agent platform — it can execute real tasks on your computer:
- Run shell commands and scripts
- Read, write, and edit files on your filesystem
- Control a browser (navigate, click, fill forms, screenshot)
- Schedule cron jobs and respond to webhooks
- Manage email, calendars, and workflows
- Automate anything you can describe in natural language
The Story Behind the Name
OpenClaw holds what may be the fastest triple rebrand in open-source history:
| Date | Name | What Happened |
|---|---|---|
| November 2025 | Clawdbot | Original name — a pun on Anthropic's "Claude" with a lobster mascot |
| January 27, 2026 | Moltbot | Renamed after Anthropic's legal team flagged trademark similarity |
| January 29, 2026 | OpenClaw | Final name — "Moltbot" never stuck, and crypto scammers hijacked old accounts |
The project was created by Peter Steinberger (founder of PSPDFKit) and went viral in late January 2026, jumping from a personal project to 185,000+ GitHub stars in weeks.
OpenClaw vs Cloud AI Assistants
| Feature | ChatGPT / Claude Web | OpenClaw |
|---|---|---|
| Where it runs | Cloud servers | Your own device |
| Data privacy | Data sent to provider | Data stays local |
| Interface | Browser tab | Your messaging apps |
| Can execute commands | No | Yes — shell, files, browser |
| Can automate workflows | Limited | Full cron, webhooks, multi-step |
| Cost | Subscription fee | Free (MIT license) + LLM API costs |
| Extensible | No | 5,700+ community skills |
| Offline capable | No | Yes (with local models via Ollama) |
Architecture: How OpenClaw Works
OpenClaw uses a Gateway architecture — a local WebSocket-based control plane that coordinates everything:
┌─────────────────────────────────────────────────┐
│ Gateway │
│ ws://127.0.0.1:18789 │
│ │
│ ┌──────────┐ ┌──────────┐ ┌──────────┐ │
│ │ Sessions │ │ Channels │ │ Tools │ │
│ └──────────┘ └──────────┘ └──────────┘ │
│ ┌──────────┐ ┌──────────┐ ┌──────────┐ │
│ │ Events │ │ Cron │ │ Skills │ │
│ └──────────┘ └──────────┘ └──────────┘ │
└────────┬──────────┬───────────┬─────────────────┘
│ │ │
┌────▼────┐ ┌──▼────┐ ┌───▼──────┐
│WhatsApp │ │ Slack │ │ WebChat │ ...13+ channels
└─────────┘ └───────┘ └──────────┘
│ │ │
┌────▼──────────▼───────────▼────┐
│ AI Model (LLM) │
│ Claude / GPT-4 / Ollama │
└────────────────────────────────┘
Key Components
| Component | What It Does |
|---|---|
| Gateway | The brain — manages sessions, routes messages, coordinates tools |
| Channels | Connectors to messaging platforms (WhatsApp, Telegram, etc.) |
| Agent Runtime | The AI reasoning engine — talks to your chosen LLM |
| Tools | Built-in capabilities: shell, filesystem, browser control, cron |
| Skills | Installable plugins from ClawHub or custom-built |
| Nodes | Companion apps (macOS, iOS, Android) that provide device capabilities |
Tech Stack
- Language: TypeScript
- Runtime: Node.js 22+
- Package Manager: pnpm (preferred), npm, or bun
- Channel SDKs: Baileys (WhatsApp), grammY (Telegram), Bolt (Slack), discord.js (Discord), signal-cli (Signal)
- Browser Automation: OpenClaw-managed Chrome/Chromium via CDP
- License: MIT — fully free and open-source
Installation and Setup
Prerequisites
- Node.js 22 or newer (verify:
node --version) - macOS, Linux, or Windows (WSL2 required on Windows)
Quick Install
macOS / Linux:
curl -fsSL https://openclaw.ai/install.sh | bash
Windows (PowerShell):
iwr -useb https://openclaw.ai/install.ps1 | iex
Or via npm/pnpm:
npm install -g openclaw@latest
# or
pnpm add -g openclaw@latest
Onboarding
After installation, run the interactive setup wizard:
openclaw onboard --install-daemon
This walks you through:
- Authentication — configuring your LLM API keys
- Gateway settings — port, workspace directory
- Channel setup — linking your first messaging platform
- Daemon install — auto-start the gateway on boot
Verify Installation
# Check gateway status
openclaw gateway status
# Open the Control UI in your browser
openclaw dashboard
# Opens http://127.0.0.1:18789/
Configuration
The main config lives at ~/.openclaw/openclaw.json:
{
agent: {
// Recommended model for best results
model: "anthropic/claude-opus-4-6",
},
gateway: {
port: 18789,
},
}
Using local models (fully offline):
{
agent: {
// Run with Ollama for complete privacy
model: "ollama/llama3.1:70b",
},
}
Environment Variables
| Variable | Purpose |
|---|---|
OPENCLAW_HOME |
Sets home directory for internal path resolution |
OPENCLAW_STATE_DIR |
Overrides state directory location |
OPENCLAW_CONFIG_PATH |
Overrides config file path |
From Source (Development)
git clone https://github.com/openclaw/openclaw.git
cd openclaw
pnpm install
pnpm ui:build
pnpm build
pnpm openclaw onboard --install-daemon
# Dev mode with auto-reload:
pnpm gateway:watch
Connecting Messaging Channels
OpenClaw supports 13+ messaging platforms out of the box:
| Channel | SDK Used | Notes |
|---|---|---|
| Baileys | Most popular channel. Uses WhatsApp Web protocol. | |
| Telegram | grammY | Bot API. Create a bot via @BotFather. |
| Slack | Bolt | Workspace app integration. |
| Discord | discord.js | Bot account in your servers. |
| Google Chat | Official API | Workspace integration. |
| Signal | signal-cli | Privacy-focused. Requires linked device. |
| BlueBubbles | BlueBubbles API | iMessage on non-Apple hardware. |
| iMessage | Legacy bridge | macOS only. |
| Microsoft Teams | Bot Framework | Enterprise integration. |
| Matrix | matrix-js-sdk | Self-hosted chat protocol. |
| Zalo | Zalo API | Popular in Vietnam. |
| WebChat | Built-in | Browser-based fallback, always available. |
Multi-Channel Setup
You can connect multiple channels simultaneously. Each channel can route to different agent configurations:
{
channels: {
whatsapp: {
enabled: true,
// Personal messages go to main agent
},
slack: {
enabled: true,
// Work Slack routes to a different agent with work-specific skills
agent: {
model: "anthropic/claude-opus-4-6",
skills: ["jira-integration", "github-pr-review"],
},
},
},
}
Multi-Agent Routing
OpenClaw supports routing different channels or accounts to isolated agents with different models, skills, and permissions. Your personal WhatsApp can have full system access, while a group Slack channel runs in a sandboxed environment.
Skills and ClawHub: Extending OpenClaw
What Are Skills?
Skills are plugins that extend what OpenClaw can do. They range from simple prompt templates to complex TypeScript integrations with external APIs.
ClawHub — The Skills Registry
ClawHub (clawhub.ai) is the community skills directory with thousands of skills across categories like:
- Productivity — Email management, calendar automation, note-taking
- DevOps — Git operations, CI/CD, server monitoring, deployment
- Finance — Expense tracking, trading automation, portfolio management
- Content Creation — Social media repurposing, blog writing, image generation
- Home Automation — IoT control, smart home routines, weather-based triggers
- Communication — Auto-replies, message scheduling, translation
- Development — Code review, documentation, debugging assistance
- Data — Web scraping, data analysis, report generation
Writing a Markdown Skill
The simplest skill is just a Markdown file with instructions:
---
name: daily-standup
description: Generate a daily standup update from git commits
---
# Daily Standup Generator
Look at the git log for the last 24 hours in the current project directory.
Summarize what was accomplished, what's in progress, and any blockers.
Format the output as:
**Done:**
- [completed items]
**In Progress:**
- [current work]
**Blockers:**
- [any issues]
Writing a TypeScript Skill
For more complex logic:
import { Skill, SkillContext } from "openclaw/skills";
export default class PriceChecker extends Skill {
name = "price-checker";
description = "Check and compare prices across shopping sites";
async execute(ctx: SkillContext) {
const { query } = ctx.params;
// Use OpenClaw's built-in browser tool
const results = await ctx.browser.navigate(
`https://www.google.com/search?q=${encodeURIComponent(query)}&tbm=shop`
);
const prices = await ctx.browser.extract("price listings");
return ctx.reply(`Found ${prices.length} results for "${query}":\n${
prices.map(p => `- ${p.title}: ${p.price}`).join("\n")
}`);
}
}
Installing Skills
# From ClawHub
openclaw skill install daily-standup
# From a GitHub repo
openclaw skill install github:username/my-skill
# List installed skills
openclaw skill list
Practical Use Cases and Workflows
1. Email Inbox Management
Tell OpenClaw to process your inbox via WhatsApp:
You: "Check my email and clean up my inbox"
OpenClaw: I'll scan your inbox now...
Found 247 unread emails:
- 180 newsletters/promotions → Archived
- 12 spam → Reported and deleted
- 35 actionable emails → Here's a priority summary:
**Urgent (3):**
1. Client proposal deadline tomorrow
2. Server alert from monitoring
3. Tax document request
**This week (8):**
...
Want me to draft replies to any of these?
2. Content Repurposing
Give it a blog post and get platform-specific versions:
You: "Take my latest blog post at /posts/ai-agents.mdx and create
versions for Twitter, LinkedIn, and Instagram"
OpenClaw: Done! Here's what I prepared:
**Twitter thread (7 tweets):**
1/ AI agents are changing how we build software...
[thread continues]
**LinkedIn post:**
[professional tone, longer format]
**Instagram caption:**
[casual tone with relevant hashtags]
Shall I post them or save as drafts?
3. DevOps Automation
Manage deployments from your phone:
You: "Deploy the staging branch to production"
OpenClaw: Running pre-deploy checks...
✓ Tests passing (142/142)
✓ No security vulnerabilities
✓ Build successful
Ready to deploy main → production.
This will affect 3 services. Proceed? (yes/no)
You: "yes"
OpenClaw: Deploying...
✓ Service A: deployed (v2.4.1)
✓ Service B: deployed (v1.8.0)
✓ Service C: deployed (v3.1.2)
All health checks passing. Deployment complete.
4. Browser Automation
You: "Compare prices for AirPods Pro on Amazon, Best Buy, and Walmart"
OpenClaw: Opening browsers and checking prices...
| Store | Price | In Stock | Shipping |
|-------|-------|----------|----------|
| Amazon | $199.99 | Yes | Free (Prime) |
| Best Buy | $189.99 | Yes | Free |
| Walmart | $194.00 | Yes | Free |
Best Buy has the lowest price at $189.99.
Want me to open the checkout page?
5. Scheduled Workflows (Cron)
Set up recurring automations:
You: "Every morning at 8am, check my calendar, summarize today's
meetings, check the weather, and send me a briefing on WhatsApp"
OpenClaw: Done! I've set up a daily cron job:
- Trigger: 8:00 AM daily
- Actions: Calendar check → Weather → Meeting prep → Send briefing
- Channel: WhatsApp
Tomorrow's briefing will be your first one.
6. Private Document Assistant (Fully Offline)
With Ollama, your data never leaves your machine:
{
agent: {
model: "ollama/llama3.1:70b",
},
}
You: "Summarize the contract at ~/Documents/vendor-agreement.pdf
and flag any concerning clauses"
OpenClaw: [processes locally with Ollama — no cloud API calls]
Summary of Vendor Agreement:
- Term: 24 months, auto-renewal
- Payment: Net 30
⚠️ Concerning clauses:
1. Section 4.2: Non-compete extends 2 years post-termination (unusual)
2. Section 7.1: Unlimited liability for data breaches (no cap)
3. Section 9.3: Vendor can change pricing with 30 days notice
Recommendation: Negotiate items 1 and 2 before signing.
Security, Privacy, and Sandboxing
Privacy Model
OpenClaw is local-first by design:
| Aspect | How It Works |
|---|---|
| Data storage | All data stays on your device |
| LLM calls | Your API key, direct to the provider (or fully local with Ollama) |
| No cloud service | No OpenClaw servers — the Gateway runs on your machine |
| No telemetry | Open-source, auditable, MIT licensed |
DM Pairing and Access Control
When someone messages your connected accounts, OpenClaw doesn't automatically respond to strangers:
- Pairing mode (default): Unknown senders must provide a pairing code you set
- Allowlist: Explicitly approve specific contacts
- Block: Reject all unknown senders
Sandboxing
For non-main sessions (group chats, shared channels), OpenClaw supports Docker sandboxing:
{
agents: {
defaults: {
sandbox: {
mode: "non-main",
// Main (personal) sessions run on host
// Group/channel sessions run in Docker containers
},
},
},
}
This means your personal WhatsApp messages get full system access, while a shared Slack channel is isolated in a container with limited permissions.
Security Audit
Run the built-in security checker:
openclaw doctor
This scans your configuration for:
- Exposed ports or endpoints
- Missing authentication
- Overly permissive tool access
- Insecure channel configurations
Advanced Configuration and Production Tips
Running on a Remote Server
You can run the Gateway on a VPS or cloud VM while companion apps on your devices handle local actions:
┌─────────────────────┐ ┌──────────────┐
│ Remote Server (VPS) │ │ Your Mac │
│ ┌───────────────┐ │ │ ┌────────┐ │
│ │ Gateway │◄─┼─────┼──│ macOS │ │
│ │ + Agent │ │ SSH/ │ │ Node │ │
│ └───────────────┘ │ Tail │ └────────┘ │
└─────────────────────┘ scale└──────────────┘
┌──────────────┐
│ Your iPhone │
│ ┌────────┐ │
│ │ iOS │ │
│ │ Node │ │
│ └────────┘ │
└──────────────┘
Remote access options:
- Tailscale Serve/Funnel: Recommended — secure, zero-config networking
- SSH tunnel: Traditional approach, requires port forwarding
- Docker: Run the entire Gateway in a container
Docker Deployment
docker run -d \
--name openclaw \
-p 18789:18789 \
-v ~/.openclaw:/root/.openclaw \
openclaw/openclaw:latest
Multi-Agent Routing
Run different agent personalities for different contexts:
{
agents: {
personal: {
model: "anthropic/claude-opus-4-6",
skills: ["email-manager", "calendar", "meal-planner"],
channels: ["whatsapp"],
},
work: {
model: "anthropic/claude-opus-4-6",
skills: ["jira", "github-pr", "slack-standup"],
channels: ["slack"],
sandbox: { mode: "always" },
},
family: {
model: "ollama/llama3.1:8b",
skills: ["recipes", "shopping-list"],
channels: ["telegram"],
},
},
}
Voice Wake and Talk Mode
On macOS, iOS, and Android, OpenClaw supports always-on voice interaction:
- Voice Wake: Say the wake word to activate (uses ElevenLabs)
- Talk Mode: Continuous voice conversation, hands-free
- Push-to-Talk: Hold a key/button to speak (macOS menu bar app)
Live Canvas
The Live Canvas is an agent-driven visual workspace (A2UI — Agent-to-UI). The AI can create and manipulate visual interfaces on your device, going beyond text-only responses.
Performance Tips
| Tip | Why |
|---|---|
| Use pnpm over npm | Faster installs, better disk usage |
| Pin your Node.js version | Avoids compatibility issues (use nvm) |
| Run Gateway as a daemon | Auto-restarts on crash, starts on boot |
| Use Tailscale for remote access | Zero-config, encrypted, no port forwarding |
| Limit tools for group sessions | Reduces attack surface |
| Enable Docker sandboxing | Isolates untrusted sessions |
Useful Commands
# Gateway management
openclaw gateway --port 18789 --verbose # Start in foreground
openclaw gateway status # Check if running
openclaw dashboard # Open Control UI
# Messaging
openclaw message send --target +1234567890 --message "Hello"
# Agent interaction
openclaw agent --message "Ship checklist" --thinking high
# Skills
openclaw skill install <skill-name>
openclaw skill list
openclaw skill remove <skill-name>
# Security
openclaw doctor # Audit configuration
# Diagnostics
openclaw logs # View recent logs
openclaw version # Check version
Getting Started
Ready to set up your own AI assistant? Here's the path:
- Install OpenClaw: Run the one-liner install script for your platform
- Run onboarding:
openclaw onboard --install-daemonwalks you through everything - Connect WhatsApp or Telegram: Start with one channel you use daily
- Try basic commands: Ask it to check your calendar, summarize a file, or search the web
- Install skills: Browse clawhub.ai and install skills for your workflows
- Set up automations: Create cron jobs for recurring tasks
- Add more channels: Connect Slack for work, Discord for communities
- Go advanced: Set up multi-agent routing, Docker sandboxing, or remote server deployment
OpenClaw turns your messaging apps into a command center. Instead of switching between a dozen browser tabs and tools, you type what you need in WhatsApp and your AI assistant handles the rest — on your hardware, with your data, under your control.