# Maestro-Playbooks

A collection of playbooks for the RunMaestro.ai agent orchestrator: saved Auto Run configurations that step an AI agent through a sequence of Markdown task documents until an exit condition is met.

Each playbook in this collection is a sequence of Markdown documents with task checkboxes that Maestro processes through AI agents, looping until a defined exit condition, such as all documents being completed or no pending items remaining. The playbooks are grouped into three categories: Assistants playbooks are one-shot setup workflows that install personal-AI frameworks such as a Life Operating System, an iMessage bridge, or an Apple Voice Memos journal onto a fresh agent.

Development playbooks work with Maestro's default agent prompt and cover tasks such as comparing two competing pull requests, raising documentation coverage to a target percentage, making a React site mobile-friendly, finding and fixing performance issues, refactoring to reduce duplication, auditing and remediating security vulnerabilities, and raising test coverage. Research playbooks require a custom agent prompt with user configuration and cover turning a source list into an interlinked knowledge base or building an Obsidian-style knowledge vault.

It is written in Python and released under the AGPL-3.0 license.

Website: https://github.com/RunMaestro/Maestro-Playbooks