AI Maestro CLI
A local dashboard for running many agents across machines: memory, code graph and agent-to-agent messaging
- 839 ★ on the repository
- Last commit
- MIT
Broad access
Why this level
- Runs agents with terminal and repository access
- Opens network connections between machines
- Installs via a script from the internet
Install
- Works in
Manual install
curl -fsSL https://raw.githubusercontent.com/23blocks-OS/ai-maestro/main/scripts/remote-install.sh | shInstalls the dashboard, AMP messaging and the Claude Code plugin. The dashboard opens at http://localhost:23000.
Access and checks
How we reviewBroad accessThe tool can change data, run commands or handle secrets.
- Runs agents with terminal and repository access
- Opens network connections between machines
- Installs via a script from the internet
This is third-party code. Review the repository files before installing.
What it does
AI Maestro runs terminal agents such as Claude Code, Codex, Aider or Cursor in tmux sessions and shows them in one web dashboard. Each agent gets persistent memory, a code index and an inbox: through the AMP protocol agents message each other directly without a human relaying context. Machines form a peer mesh with no central server, and an agent can move to another computer along with its repositories. The installer also adds a Claude Code plugin with skills and CLI scripts for memory search, code graph queries and messaging.
Who it is for. For developers and small teams running several agents at once on one or more machines.
Good fit when
- Three or more agents run at once and you relay context by hand
- Agents live on different computers
- You need agent memory that survives sessions
Not a fit when
- You use one agent in one terminal
- You cannot run a persistent service on the machine
Example request
Message the backend agent about the new API field and ask it to confirmLimitations
Requires Node.js 18+ and tmux; Windows only via WSL2. Plugin skills work together with the running AI Maestro service. Agent subscriptions are paid separately. Installs via a script from the internet.
How to disable. Stop the AI Maestro service, uninstall its Claude Code plugin via /plugin uninstall and delete the install directory.
FAQ
Do I have to use Claude Code?
No, any terminal agent works: Codex, Aider, Cursor and others.
Is there a paid tier?
No, it is MIT licensed with no accounts or telemetry; you only pay for the agents themselves.
Discussion
No questions or reviews yet.
Related
CLI toolsA tmux-based TUI that runs Claude Code, Codex and other agents in one list: status, quick prompts, worktrees and diff review
Alook gives local Claude Code, Codex, Cursor and OpenCode agents a handle and an inbox so a team can talk to them in shared rooms like to people
A portable .agent folder with memory, skills and protocols that plugs into many coding agents and keeps its knowledge when you switch tools
A control plane that runs Claude Code, Codex and Gemini CLI agents in parallel with a shared task board and schedules
A CLI that keeps coding agents' shared context in one place: instructions, skills, MCPs, subagents and durable memory across projects and machines
CLI Agent Orchestrator
CLI Agent Orchestrator (CAO)
A CLI that coordinates several coding agents: a supervisor delegates tasks to specialists in parallel or in sequence across isolated tmux sessions
Comparison with similar tools
Entries that share the main search query, and how they differ.
README in short
The README tells how the author got tired of relaying context between dozens of agents and describes a model where an agent owns a product or repository and works through git like an employee. It gives a one-command install, a manual yarn setup and Windows and Linux notes. It then covers the dashboard, memory, messaging and multi-machine features. MIT licensed.