Open source · MIT · 376k★ on GitHub

OpenClaw

Your own personal AI assistant. Any OS. Any platform. An open-source, local-first agent that connects 23+ messaging channels to AI that executes real tasks — managed or self-hosted.

or read the install guide →

What is OpenClaw?

OpenClaw is a local-first, open-source personal AI assistant built as a long-running Node.js service. It bridges every messaging platform you already use — WhatsApp, Telegram, Slack, Discord, iMessage and 18 more — to a single AI agent that can browse the web, run tools, and automate workflows on your behalf.

Because it runs as a persistent daemon on your own hardware (or a VPS), your agent is always available to receive messages, execute scheduled tasks, and respond from any channel — 24/7, without keeping a tab open. You bring your own LLM provider key (BYOK): Claude 4, GPT-4o, Gemini 2.0, DeepSeek V3 and other major models are all supported.

Everything you need

Multi-channel inbox

23+ messaging platforms unified into a single AI agent — WhatsApp, Telegram, Slack, Discord and more.

Voice wake-word & talk mode

Hands-free voice interaction on macOS, iOS, and Android with always-on wake-word detection.

Live Canvas visual workspace

A shared visual workspace for rendering results, images, and structured outputs alongside the chat.

First-class browser & tool use

Built-in browsing, web search, file operations, and extensible tool integrations out of the box.

Multi-agent routing

Route tasks across multiple isolated agent workspaces, keeping personal and work contexts separate.

Companion apps

Native companion apps for macOS, iOS, and Android for seamless on-device access.

Skill management & onboarding

Install, manage, and configure agent skills through an interactive onboarding wizard.

Cron automations & webhooks

Schedule recurring tasks, trigger workflows via webhooks, and listen for Gmail Pub/Sub events.

SSH/Docker deploy

Deploy remotely over SSH or containerise with Docker for flexible infrastructure options.

Channels & models

Supported messaging channels (23+)

WhatsAppTelegramSlackDiscordGoogle ChatSignaliMessageMicrosoft TeamsMatrixIRCFeishuLINEMattermostNextcloud TalkNostrSynology ChatTwitchWeChatQQWebChat+ more

LLM providers (BYOK)

Claude 4GPT-4oGemini 2.0DeepSeek V3OpenRouter (200+ models)Any BYOK provider

Your API key is yours — OpenClaw never stores it.

Run it your way

Self-host

Install OpenClaw on any machine you control. You manage updates, backups, and monitoring.

Requirements

  • Node.js 22.19+ or 24
  • macOS, Linux, or Windows WSL2
  • Your LLM provider API key (BYOK)
npm install -g openclaw@latest
openclaw onboard --install-daemon

Need production setup (HTTPS, process manager, reverse proxy)? Read the full self-host guide →

Recommended

Managed by AgentOcean

Skip the setup — run managed OpenClaw on a dedicated VPS in ~30 seconds, with HTTPS, daily backups and updates handled for you.

  • Live in ~30 seconds
  • Dedicated VPS — isolated from other users
  • Automatic updates & daily backups
  • HTTPS + custom domain support
  • 24/7 uptime monitoring
  • Your API key never touches our servers

Also available: a cheaper self-managed optimized-VPS option — same pre-tuned server, you handle the agent.

View managed OpenClaw plans

Frequently asked questions

What is OpenClaw?

OpenClaw is a local-first, open-source personal AI assistant. It is a long-running Node.js service that connects 23+ messaging channels you already use — WhatsApp, Telegram, Slack, Discord and more — to an AI agent that executes real tasks: browsing the web, running tools, and automating workflows.

Is OpenClaw free and open source?

Yes. OpenClaw is fully open source under the MIT license. You can download it, modify it, and self-host it for free. The GitHub repository is at github.com/openclaw/openclaw.

Which messaging apps does OpenClaw support?

OpenClaw supports 23+ platforms including WhatsApp, Telegram, Slack, Discord, Google Chat, Signal, iMessage, Microsoft Teams, Matrix, IRC, Feishu, LINE, Mattermost, Nextcloud Talk, Nostr, Synology Chat, Twitch, WeChat, QQ, and WebChat.

What does it cost to run OpenClaw?

OpenClaw is free and open source (MIT). You bring your own AI provider API key (BYOK) — it works with Claude 4, GPT-4o, Gemini 2.0, DeepSeek V3 and other major LLMs. You then need a place to run it: a small $5–$10 VPS if you self-host, or self-managed hosting from $9.99/mo and fully managed from $13.99/mo via AgentOcean.

Can AgentOcean host OpenClaw for me?

Yes. AgentOcean offers fully managed OpenClaw hosting on a dedicated VPS. We handle provisioning, HTTPS, daily backups, and automatic updates — your agent is live in about 30 seconds. There is also a cheaper self-managed optimized-VPS option where you run the agent yourself.

How do I install OpenClaw myself?

You need Node.js 22.19+ (or 24) on macOS, Linux, or Windows WSL2. Install via npm: npm install -g openclaw@latest then run openclaw onboard --install-daemon. Alternatively, clone from source and run npm install && npm start.

Also hosting Hermes Agent — the self-improving AI agent by Nous Research with a built-in learning loop and persistent memory.

Ready to launch your OpenClaw agent?

Skip the setup headache. AgentOcean provisions a dedicated VPS with OpenClaw installed, HTTPS configured, and daily backups running — in about 30 seconds.