# AgentsRoom > Your IDE for Claude agents. The visual alternative to CMUX for running multiple Claude agents. AgentsRoom is a desktop app (macOS) with a mobile companion (iOS/Android) that lets you run dozens of Claude CLI agents across all your projects at once. Think WhatsApp, but for your AI dev team. ## What it does - **Visual project management**: All your projects organized by zones (Work, Personal, Side-projects) with live activity indicators - **Multi-agent orchestration**: Assign specialized roles to agents across your projects (14 built-in roles) - **Real-time terminal per agent**: Full xterm.js terminal with syntax colors, scrollback, and live streaming output - **Agent status tracking**: See at a glance who's coding, thinking, done, idle, or needs input - **macOS notifications + mobile push**: Get notified on desktop and phone when an agent finishes or needs your input - **CLAUDE.md editor**: Edit project context, permissions, and hooks without leaving the app - **Model per agent**: Choose Opus, Sonnet, or Haiku per agent depending on the task - **Prompt library**: Per-project and global prompt collections with folders and tags, synced to the cloud - **Git integration**: Status, diff, commit, pull, and AI-generated commit messages from the UI - **Cloudflare Tunnel**: Share local dev servers as public HTTPS URLs for live mobile preview - **Usage tracking**: Monitor token consumption (in/out/cache) and rate limits per agent - **Mobile companion**: Monitor agents, read chat-style output, get push notifications, and preview dev servers from your phone - **End-to-end encryption**: Desktop-to-mobile sync uses X25519 + XSalsa20-Poly1305 (E2EE) - **Dev Terminals**: Built-in process manager to launch all project services (backend, frontend, mobile) with one click, with live status tracking and AI command generation - **100% local execution**: Agents run on your machine via Claude CLI. Your code never leaves your disk. ## How it works 1. Create a project pointing to a real folder on your machine 2. Add agents with specialized roles (each gets its own system prompt and focus areas) 3. Send prompts. AgentsRoom spawns real Claude CLI processes. 4. Watch the output stream live, get notified when done, give new instructions 5. Monitor progress from your phone via the companion app ## Tech stack - **Desktop**: Electron 33 + React 19 + Vite 6 + Zustand 5 + xterm.js + Tailwind CSS 3 - **Mobile**: Expo 55 + React Native 0.83 + Zustand 5 - **Site**: Next.js 16 + React 19 + Prisma 7 + MariaDB + Tailwind CSS 4 - **Relay**: Node.js + ws (WebSocket relay for E2EE desktop-mobile sync) - **Agent runtime**: Claude CLI (real process per agent) ## Comparison with alternatives | Feature | AgentsRoom | CMUX | AMUX | iTerm2 | tmux | |---------|-----------|------|------|--------|------| | Visual GUI | Yes | No | Web UI | Yes | No | | Agent status | Yes | No | Basic | No | No | | Notifications | Desktop + Mobile | CLI only | No | No | No | | Multi-project view | Yes | No | No | Manual | Manual | | Built-in terminal | Yes | Yes | Yes | Yes | Yes | | Edit CLAUDE.md | Yes | No | No | No | No | | Mobile companion | Yes | No | No | No | No | | Token tracking | Yes | No | Basic | No | No | | Drag and drop | Yes | No | No | No | No | | Setup time | 30 sec | 5 min | 10 min | Manual | Manual | ## 14 built-in agent roles 1. **DevOps** - CI/CD, infra, deployment. Example: "Set up CI to run tests on every PR" 2. **Fullstack** - End-to-end features. Example: "Add Stripe checkout flow with webhook handling" 3. **Frontend** - UI/UX, components. Example: "Refactor the Header with the new design tokens" 4. **Backend** - API, database, performance. Example: "Add rate limiting to the auth endpoints" 5. **Architect** - System design, refactoring. Example: "Split the monolith into 3 bounded contexts" 6. **QA** - Tests, edge cases, quality. Example: "Write unit tests for the payment module" 7. **Marketing** - Copy, SEO, landing pages. Example: "Rewrite the pricing page copy for conversion" 8. **PM** - Specs, priorities, coordination. Example: "Write the spec for the notification system" 9. **Security** - Audit, vulnerabilities. Example: "Audit the auth flow for OWASP top 10" 10. **Mobile** - iOS, Android, React Native. Example: "Add offline support with local SQLite cache" 11. **Git** - Version control, branching, merges. Example: "Rebase feature branch and resolve conflicts" 12. **SEO** - Search optimization, structured data. Example: "Add JSON-LD schema to all product pages" 13. **i18n** - Localization, translations. Example: "Translate the app to Japanese and Korean" 14. **Custom** - Any role you define with a custom system prompt ## Features ### Agent Notifications https://agentsroom.dev/features/agent-notifications Unread notifications and status alerts for AI agents. Includes unread badges per session, WhatsApp-style message previews, desktop notifications, macOS dock badges, and mobile push notifications. Know when an agent finishes a task or needs your input without watching the terminal. ### Prompt Library https://agentsroom.dev/features/prompt-library Built-in prompt manager for saving, organizing, and reusing Claude prompts per project. Prompts are stored as JSON files in `.agentsroom/prompts.json` (git-committed) with personal prompts gitignored. Supports drag-and-drop organization, cloud backup, and one-click sending to agents. ### Agent Status Tracking https://agentsroom.dev/features/agent-status-tracking Real-time monitoring dashboard showing individual agent statuses: Thinking, Done, Needs Input, Idle. Includes project-level indicators with counts of working and completed agents, hover previews, and sidebar badges to track multiple agents across all your projects. ### Vibe Coding Review https://agentsroom.dev/features/vibe-coding-review Side-by-side diff viewer for reviewing AI-generated code changes. Features word-level diffs, syntax highlighting, keyboard navigation, file list with color-coded status (added, deleted, modified), ability to ask Claude about specific files, and AI-generated commit messages. ### Per-Agent Review https://agentsroom.dev/features/per-agent-review Filtered code review interface that isolates changes by individual AI agent. Shows per-agent tabs, file tracking to see which agent modified what, scoped commits per agent, filtered keyboard navigation, and attribution to track which agent made which modifications. ### Mobile-Desktop Sync https://agentsroom.dev/features/mobile-desktop-sync Real-time encrypted sync between the desktop app and the mobile companion (iOS and Android). Start Claude Code agents from your phone, type in the terminal remotely, send saved prompts, preview dev servers through Cloudflare tunnels, and monitor agent status with push notifications. All communication uses X25519 key exchange and XSalsa20-Poly1305 encryption. ### Dev Terminals https://agentsroom.dev/features/dev-terminals Built-in terminal and process manager for launching all project services from one dashboard. Define start commands for your backend, frontend, mobile app, and microservices, then launch them all with one click. Includes live process status tracking (running, starting, stopped, crashed), command snippets for frequent tasks, AI-powered command generation via Claude that scans your project structure, 30+ framework category icons (React, Next.js, Django, Rails, Go, Docker, etc.), and git-native storage in .agentsroom/commands.json. ## Internationalization The site is available in 10 languages: English, French, Spanish, Portuguese, German, Russian, Japanese, Korean, Chinese (Simplified), and Polish. ## SEO content strategy The site targets keywords such as: - "IDE for Claude", "Claude IDE", "Claude agent manager" - "CMUX alternative", "AMUX alternative", "Claude multiplexer" - "multi-agent Claude", "run multiple Claude agents" - "AI coding swarm", "parallel AI coding", "vibe coding" - "Claude Code tips", "CLAUDE.md guide", "CLAUDE.md best practices" - "AI dev team", "AI development team", "solo developer AI agents" - "monitor AI agents", "AI agent dashboard", "AI coding monitor" - "Cursor alternative", "AI code review", "parallel coding" - "Claude system prompt", "system prompt for Claude developer roles"