Manage Claude Code, Aider, Cursor, and any terminal-based AI agent from a beautiful web interface. Built for developers who love AI pair programming.
Your AI agents are scattered across terminals, computers, and cloud servers. They forget everything between sessions. They can't talk to each other. You're the bottleneck - copy-pasting context, relaying messages, and manually coordinating their work.
AI Maestro is an AI Agent Orchestrator that gives your agents superpowers. Persistent memory that grows over time. Direct agent-to-agent communication. Run agents on your laptop, remote servers, or Docker containers - all from one dashboard. Your agents become a coordinated team, not isolated tools.
Your AI agents now have persistent memory. Code Graph visualization, background subconscious processing, and semantic search across all conversations.
Works with any terminal-based AI: Claude Code, Aider, Cursor, GitHub Copilot CLI, OpenAI Codex, and your custom AI scripts.
3-level hierarchy using hyphens (e.g., project-category-agent). Dynamic color coding and auto-grouping for instant visual organization.
WebSocket streaming for real-time terminal I/O. No lag, no polling. Keyboard shortcuts for power users.
Take notes for each agent, auto-saved to localStorage. Track decisions, save commands, keep TODOs.
Your AI agents can now talk to each other. Built-in inbox/outbox with unread message tracking. See only new messages, mark as read automatically, and never miss important updates.
Monitor your AI agents from your phone or tablet. Touch-optimized interface with secure Tailscale connectivity.
Control AI agents across multiple computers and cloud servers from one dashboard. Scale from local to cloud seamlessly.
Browse, search, and analyze every conversation your agents have had. Full thinking messages and tool usage tracking.
Code Graph visualization, persistent memory with CozoDB, background subconscious processing, and semantic search across all conversations.
Create new agents, export to .zip files, import on any machine, and transfer between hosts. Full lifecycle management for your AI agents.
Auto-discovers all tmux sessions instantly. No config files, no manual setup. Just run and start coding.
MIT licensed. No subscriptions, no hidden costs. Built by developers, for developers. Community-driven and transparent.
From simple local setups to enterprise distributed architectures. Start where you are, scale when you need to.
Beginner • 5 min setup
Run multiple AI agents on one machine. Perfect for getting started and solo development.
Intermediate • 15 min setup
Control agents across laptop, desktop, and servers. Distribute workload intelligently.
Intermediate • 10 min setup
Containerized agents with perfect isolation. Zero dependency conflicts.
Advanced • 20 min setup
Local + cloud containers. Best of both worlds with unlimited scale.
Expert • 30 min setup
Everything combined. Enterprise-grade orchestration at maximum power.
macOS / Linux:
git clone https://github.com/23blocks-OS/ai-maestro.git
cd ai-maestro
yarn install
yarn dev
Windows (WSL2):
# 1. Install WSL2 (PowerShell as Administrator)
wsl --install
# 2. Restart Windows, then in Ubuntu terminal:
git clone https://github.com/23blocks-OS/ai-maestro.git
cd ai-maestro
yarn install
yarn dev
# 3. Access from Windows browser:
Dashboard opens at http://localhost:23000
From the UI (Recommended):
myproject-backend-api → creates 3 levelsproject-backend → 2 levelsproject-backend-api → 3 levelsYour agent is now live in the dashboard. Click to switch between agents. Add notes. Organize your work. That's it.
Every agent has persistent memory powered by CozoDB. They learn your codebase and remember past conversations.
Direct agent-to-agent messaging. Your frontend agent can request APIs from backend agent without you playing messenger.
Run agents on your laptop, remote servers, Docker containers, or cloud VMs. Manage them all from one dashboard.
Code Graph visualization shows your entire codebase structure. Agents know what files relate to what before they even start.