RunMaestro/Maestro

Agent Orchestration Command Center

What it solves

Maestro is a cross-platform desktop application designed for power users and hackers who manage multiple AI coding agents and projects simultaneously. It eliminates the fractured attention caused by juggling multiple terminal sessions and provides a high-velocity, keyboard-first interface for orchestrating a fleet of agents.

How it works

Maestro acts as a pass-through to existing AI providers and coding agents (such as Claude Code, OpenAI Codex, OpenCode, and Copilot-CLI). It leverages the existing MCP tools, permissions, and authentication of these providers. Instead of interactive sessions, Maestro manages tasks as prompts and responses, allowing for parallel execution and session discovery.

Who it’s for

It is built for developers and power users who prefer keyboard-driven workflows and want to run multiple AI agents in parallel across different git branches or projects.

Highlights

  • Git Worktrees: Run agents on isolated branches in separate directories to enable parallel development without conflicts.
  • Auto Run & Playbooks: Batch-process markdown checklists as repeatable workflows, where each task is executed in a fresh AI session for clean context.
  • Group Chat: Coordinate multiple AI agents in a single conversation moderated by an AI orchestrator.
  • Keyboard-First Design: Extensive customizable shortcuts and a Cmd+K quick actions palette for rapid navigation.
  • Mobile Remote Control: Monitor and control agents via a built-in web server with QR code access.
  • Integrated Tooling: Includes a file explorer with @ mentions, a git diff viewer, and a usage dashboard for token and cost tracking.