Teach Maestro

This is your project onboarding interview for AI workflows. It walks you through five sections (models, architecture, evaluation, constraints, priorities) and generates a `.maestro.md` context file that other Maestro commands depend on. Think of it as the setup wizard that asks about your stack, budget, latency needs, and what you actually care about optimizing for. Run it once per project before anything else. The structured approach is solid, though you'll need to be specific in your answers or the downstream commands won't have much to work with. It won't let you skip sections or overwrite existing context without confirmation, which is the right call.

skills.sh installs: 352
GitHub stars: 602
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add sharpdeveye/maestro --skill teach-maestro --agent claude-code

Installs into .claude/skills of the current project.

Files

BOOTSTRAP — First Command to Run

This is the entry point for Maestro. It creates the .maestro.md context file that all other Maestro commands depend on. No other preparation is needed — this IS the preparation.

You are conducting a structured interview to understand this project's AI workflow setup. Be conversational but thorough.

Interview Questions

Ask these questions one section at a time. Wait for answers before proceeding.

Section 1 — Models & Providers

What AI model(s) are you using? (e.g., GPT-4, Claude, Gemini, local models)
Are you using APIs directly or through a framework? (e.g., LangChain, LlamaIndex, custom)
What are your context window sizes?

Section 2 — Workflow Architecture