OpenClaw: Complete Guide – Self-Hosted AI Assistant for WhatsApp, Telegram & More

https://openclaw.ai

OpenClaw is an open-source personal AI assistant that runs on your own machine or server. It connects the chat apps you already use—WhatsApp, Telegram, Discord, Slack, iMessage, and more—to AI coding agents. Your assistant, your machine, your rules.

What is OpenClaw?

OpenClaw is a self-hosted gateway that connects your favorite chat apps to AI agents like Pi. You run a single Gateway process on your own machine (or a server), and it becomes the bridge between your messaging apps and an always-available AI assistant. Unlike SaaS assistants where your data lives on someone else's servers, OpenClaw runs where you choose: laptop, homelab, or VPS. Your infrastructure, your keys, your data.

Originally created by developer Peter Steinberger (first as Clawd, then Moltbot), the project was renamed to OpenClaw and has grown to over 100,000 GitHub stars. It is MIT licensed and community-driven.

Key Features

  • Multi-channel gateway: WhatsApp, Telegram, Discord, iMessage, Slack, and more with a single Gateway process
  • Plugin channels: Add Mattermost, Twitch, Google Chat, and others via extension packages
  • Self-hosted: Runs on your hardware—no data leaves your control
  • Agent-native: Built for coding agents with tool use, sessions, memory, and multi-agent routing
  • Media support: Send and receive images, audio, and documents
  • Web Control UI: Browser dashboard for chat, config, sessions, and nodes
  • Mobile nodes: Pair iOS and Android nodes with Canvas support
  • Open source: MIT licensed, transparent, and community-driven

How It Works

The Gateway is the single source of truth for sessions, routing, and channel connections. Chat apps and plugins connect to the Gateway; the Gateway routes messages to your AI agent (e.g. Pi) and back. You can also use the CLI, Web Control UI, or macOS app to interact with the same assistant.

OpenClaw is not an AI model itself—it is a runtime and message router that connects to AI models (e.g. Claude, GPT-4, Gemini), giving them the ability to work across your chat apps and perform actions.

Requirements

  • Node.js 22+
  • API key: Anthropic (recommended), OpenAI, or other supported providers
  • OS: macOS, Linux, or Windows (via WSL2)
  • Cost: OpenClaw itself is free. You pay for your server (if any) and API usage (typically around $5–50/month depending on usage)

Quick Start

Step 1: Install OpenClaw

npm install -g openclaw@latest

Step 2: Onboard and install the service

openclaw onboard --install-daemon

Use the guided setup and pairing flows as prompted.

Step 3: Pair a channel and start the Gateway

openclaw channels login
openclaw gateway --port 18789

After the Gateway starts, open the Control UI in your browser at http://127.0.0.1:18789/ (or use Tailscale for remote access).

Configuration

Config lives at ~/.openclaw/openclaw.json. You can lock down who can use the assistant (e.g. channels.whatsapp.allowFrom) and set mention rules for groups. If you do nothing, OpenClaw uses the bundled Pi binary in RPC mode with per-sender sessions.

Security

OpenClaw has invested in security hardening and provides security best practices in the docs. Prompt injection remains an industry-wide challenge; use strong models and follow the recommended practices. Keep your API keys and config secure.

Get Started and Learn More

Whether you want a private AI assistant in your pocket or a self-hosted agent for your team, OpenClaw gives you full control. Welcome to OpenClaw. 🦞

← AI Tools