Maestro Setup bootstraps a repository into a long-running agent harness. It inspects manifests and repo files, copies opinionated style guides, creates .maestro/context/ docs, injects or updates a root AGENTS.md block, and writes a .maestro/setup-report.md detailing actions, detected languages, and TODOs. The skill is non-interactive and idempotent: re-running it is safe and will not overwrite user content outside managed markers.
Use Maestro Setup when you want to prepare a codebase for agent-driven work — for example, to: initialize project knowledge (AGENTS.md), standardize code style guides, add a lightweight CI check, produce an evidence-first setup report, or create the .maestro/context/ docs that downstream agent skills will rely on. Ideal during onboarding, repository handoffs, or before starting larger agent-driven design tasks.
reference/ bundled materials when present.maestro/setup-report.md. The skill documents marker usage and non-destructive update rules.This skill is intended for assistants that can read repository files and emit filesystem changes (CLI-capable agents like Claude Code, Codex, Copilot). It is non-interactive and expects a runtime that can write files and create workflows.
This skill has not been reviewed by our automated audit pipeline yet.
