UniFi Design vs ClawTalk

UniFi Design brings the full UniFi Design Center experience to your device, purpose-built for planning, building, and visualizing your setup from anywhere. Design your setup with the Wizard, now with far more control. Add and configure devices, customize your rack down to the detail, and sync your projects directly from UniFi Design Center so your work moves with you. - Build and edit complete rack setups with detailed configuration options. - Sync projects and floor plans from UniFi Design Center. - Visualize your network in a redesigned topology view to see your entire project at a glance. - Plan every connection with a dedicated ports view to edit devices and ports directly. - Place a full 1:1 rack in your room or office using AR to see exactly how it will look. - Access Certified Installers, Network Ports AR, Tech Specs, WiFi Calculator, and Capacity Calculator in one place. From idea to deployment, UniFi Design gives you everything you need to plan your project, wherever you are.

Have a real, hands-free conversation with your AI agents. ClawTalk is a native iOS client for OpenClaw, the open-source AI agent framework — and it's built around voice. CONVERSATION MODE Tap once and start talking. ClawTalk listens for the end of your sentence, transcribes it on-device with WhisperKit on the Apple Neural Engine, sends it to your agent, streams the reply, and speaks it back to you. Then it starts listening again — automatically. No buttons, no taps between turns. iOS voice processing (the same audio pipeline Zoom and FaceTime use) handles echo cancellation, automatic gain, and noise suppression, so the agent's own voice never re-triggers the mic. Your audio never leaves the device. Only the resulting text is sent to your gateway. PUSH-TO-TALK Prefer to just dictate a single message? Press and hold the mic. Release to send. STREAMING RESPONSES Text streams in real-time as your agent generates it. Full markdown support: code blocks, tables, lists, inline formatting. TEXT-TO-SPEECH Choose how the agent speaks back: - ElevenLabs — highest quality, most natural voices - OpenAI — great quality, cost-effective - Apple — free, works offline, no API key needed MULTI-AGENT CHANNELS Separate channels for separate agents. Each channel keeps its own conversation history. Browse the agents on your gateway, or type an agent ID manually. IMAGE SENDING Attach up to 8 photos per message. Your agent can see and respond to images. TOOLS DASHBOARD Reach into your agent's internals without going through chat — memory, sessions, available agents, browser status, models. SECURITY FIRST - Credentials live in the iOS Keychain (encrypted by the Secure Enclave) - HTTPS enforced on every connection — plain HTTP is rejected - On-device speech-to-text — audio never reaches any server - Chat history encrypted at rest with iOS Data Protection - TLS 1.2 minimum WHAT YOU NEED ClawTalk connects to your own OpenClaw gateway over HTTPS. You'll want: 1. A running OpenClaw instance with the HTTP API enabled 2. A secure path to it — Cloudflare Tunnel, Tailscale, or similar 3. A gateway token ClawTalk ships no AI models and no cloud services. It's a client for your self-hosted OpenClaw setup. The GitHub repo has setup instructions. OPEN SOURCE ClawTalk is MIT-licensed and fully open source on GitHub. Build it yourself for free, or buy from the App Store to support ongoing development and get automatic updates.