AI Maestro orchestrates multiple AI agents, providing a unified dashboard for managing them across different machines. It addresses the challenge of inter-agent communication and context loss by enabling persistent memory, agent-to-agent messaging, and a distributed peer mesh network. This empowers developers running numerous AI agents, teams coordinating AI-assisted workflows, and creators integrating agents with external services, transforming AI agents into a collaborative, productive workforce.

How It Works

The system employs a peer mesh network architecture, allowing agents and dashboards to run on any machine without a central server, facilitating distributed AI compute. Direct agent-to-agent communication is enabled via the Agent Messaging Protocol (AMP), which supports priority levels and cryptographic signatures. Persistent intelligence is built through three layers: conversational memory, an interactive Code Graph with delta indexing for codebase visualization, and auto-generated, searchable documentation.

Quick Start & Requirements

• Primary install: curl -fsSL https://raw.githubusercontent.com/23blocks-OS/ai-maestro/main/scripts/remote-install.sh | sh
• Prerequisites: Node.js 18+, tmux, build-essential (Linux). Windows users require WSL2.
• Estimated setup time: 5-10 minutes.
• Relevant links: Quick Start Guide, Full Windows guide.

Highlighted Details

• Multi-machine peer mesh network enabling distributed agent deployment without a central server.
• Agent Messaging Protocol (AMP) for seamless, email-like communication and coordination between agents.
• Persistent memory layers including conversational recall, a Code Graph for codebase analysis, and auto-generated documentation.
• Work coordination tools like agent teams, split-pane "war rooms," and a Kanban board for task management.
• Organizational gateways for secure integration with Slack, Discord, Email, and WhatsApp.

Maintenance & Community

The project is associated with Juan Pelaez and 23blocks. Contributions are welcomed via CONTRIBUTING.md. A roadmap is available for future development insights.

Licensing & Compatibility

Licensed under the MIT license, it is free for any purpose, including commercial use.

Limitations & Caveats

The project is at version 0.24.7, indicating active development. Specific limitations, known bugs, or unsupported platforms are not detailed in the provided README.