OpenClaw Releases iOS and Android Companion Node Apps That Connect a Phone to a Self-Hosted AI Agent Gateway OpenClaw released free iOS and Android companion apps that connect a phone to a self-hosted AI agent Gateway, turning the device into a node with access to camera, location, voice, and other hardware capabilities. The Gateway runs the agent on macOS, Linux, or Windows, while the phone acts as a peripheral for input and output, with privacy controls requiring explicit approval for sensitive commands. OpenClaw just released native companion apps for iOS and Android https://play.google.com/store/apps/details?id=ai.openclaw.app . The iOS app https://apps.apple.com/us/app/openclaw-ai-that-does-things/id6780396132 is listed as ‘OpenClaw – AI that does things.’ Both apps are free to download. They are not standalone chatbots. Each phone becomes a node in a self-hosted agent network. The assistant itself runs on a separate Gateway. That separation is the whole design. TL;DR - OpenClaw’s iOS and Android apps are companion nodes, not standalone assistants. - The Gateway runs the agent; phones add camera, location, voice, and Canvas. - Nodes pair over WebSocket on port 18789 and require explicit approval. - Privacy-heavy commands stay off until you allowlist them. - A Gateway on macOS, Linux, or Windows WSL2 is required. What is OpenClaw ? OpenClaw is an open-source personal AI assistant/agent. It was created by Peter Steinberger with community contributors. The project is independent and not affiliated with Anthropic. Its core is written in TypeScript. The runtime is Node 24 recommended or Node 22.19+. The Gateway runs on macOS, Linux, or Windows via WSL2. You talk to it from chat apps you already use. Supported channels include WhatsApp, Telegram, Discord, Slack, Signal, and iMessage. The agent can browse the web, run shell commands, and read and write files. It works with hosted, subscription-backed, gateway, or local models. You bring an API key from your chosen provider. It keeps persistent memory and supports community skills and plugins. How the Gateway-and-Nodes Architecture Works ? The Gateway is the single control plane. It owns sessions, routing, channels, tools, and events. You run one Gateway process on your own machine or server. Chat messages always land on the Gateway, never on a phone. A node is a companion device that connects to that Gateway. Nodes connect over a WebSocket on default port 18789. Each node registers with role: "node" during pairing. Nodes expose a command surface through node.invoke . Those command families include canvas. , camera. , device. , notifications. , and system. . The documentation is explicit:’Nodes are peripherals, not gateways.’ On a local network, apps discover the Gateway via mDNS/Bonjour. For remote access, OpenClaw recommends Tailscale with a wss:// endpoint. What does Mobile Apps Add ? The phone gives the agent a body. It grants device-specific hardware to your workflows. The iOS app pairs by QR code or setup code. It supports chat, realtime and background Talk mode, and approvals. You can share text, links, and media from iOS into OpenClaw. Optional capabilities include camera, screen, location, photos, contacts, calendar, and reminders. The Android app is a companion node, not a standalone gateway. It offers streaming chat replies, image attachments, and full session history. Talk Mode uses ElevenLabs or system TTS. A live Canvas surface lets the agent render dashboards and tools. Android grants permissions one by one. A foreground service keeps the Gateway connection alive. iOS Node vs Android Node | Capability | iOS — OpenClaw – AI that does things | Android — OpenClaw’ node | |---|---|---| | Role | Companion node | Companion node | | Pairing | QR code or setup code | Setup code or manual host/port | | Chat | Chat from iPhone | Streaming replies, image attachments, full session history | | Voice | Realtime and background Talk mode | Talk Mode ElevenLabs or system TTS | | Canvas | Canvas surface | Live Canvas surface | | Device capabilities | Camera, screen, location, photos, contacts, calendar, reminders | Camera, photos, screen capture, location, notifications, contacts, calendar, SMS, motion sensors | | Action approvals | Reviewable from the iPhone | Managed on the Gateway | | Data collection | None declared App Store | None declared Google Play | | Requirement | iOS 18.0+ and a running Gateway | A running Gateway on macOS, Linux, or Windows WSL2 | Use Cases With Examples Consider field data collection on a job site. The agent uses iOS camera capture to photograph conditions. Location tags each photo with GPS coordinates. Consider a context-aware reminder. The agent triggers a task when you reach a place. Consider an incoming notification on Android. The agent reads it and drafts a reply. Consider a live dashboard. The agent pushes a Canvas surface to your screen. Consider hands-free use. Talk Mode holds a continuous voice conversation. One caveat applies to camera and screen capture. They require the app in the foreground. Background calls return an error. Pairing a Phone: A Minimal Walk-Through First, run the Gateway on a supported host. On the Gateway host macOS, Linux, or Windows via WSL2 npm install -g openclaw@latest openclaw onboard --install-daemon Next, open the app and select a discovered Gateway. Or enter the host and port manually. The app connects with role: "node" and sends a device pairing request. Approve it from the Gateway CLI. openclaw devices list openclaw devices approve