Maestro

One conductor. An orchestra of AI agents. Package your delivery workflows as YAML, and build toward repo-native AI operating systems with governed context, lineage, and agent execution.

Get Started Explore Maestro-Prime View on GitHub

Install in seconds

curl -fsSL https://maestro.mahoosuc.ai/install | bash

irm https://maestro.mahoosuc.ai/install.ps1 | iex

brew install webemo-aaron/maestro/maestro

docker run --rm -v "$PWD:/workspace" ghcr.io/webemo-aaron/maestro:latest workflow list

What is Maestro?

Maestro is a workflow engine for AI coding agents. Define multi-step development workflows in YAML — code review, bug fixes, feature implementation, testing — and run them with a single command.

Repeatable

Package your best AI coding patterns as shareable YAML workflows

Isolated

Each workflow runs in its own git worktree — no conflicts, no mess

Portable

Run from CLI, Web UI, Slack, Telegram, GitHub, or Discord

Composable

Chain nodes into DAGs with dependencies, loops, and conditional logic

Multi-provider

Works with Claude Code SDK and Codex SDK

Build toward Maestro-Prime

Maestro is the workflow and execution substrate. Maestro-Prime extends that foundation into a repo-native A2A operating system / solution stack with governance, semantic ODD translation, deployment lineage, and audited learning.

Repo-Native Memory

Treat the repository as the canonical operating memory for product, architecture, runbooks, and agent context.

Governed Agents

Specialist agents operate through owned artifacts, review handoffs, and validation rules instead of free-form autonomy.

Operational Lineage

Track deployment baselines, configuration obligations, drift findings, and learning history in an audited control plane.