Orchestrate multi-agent coding delivery loops with Maestro Flow
Use Maestro Flow when a coding operator wants intent-driven planning, execution, verification, review, retry, and long-running multi-agent loops from one workflow surface.
npx skills add agentskillexchange/skills --skill orchestrate-multi-agent-coding-delivery-loops-with-maestro-flow
Use Maestro Flow when a repository task needs more than a single prompt-response coding run. Its Ralph and Odyssey workflows route a natural-language goal through analysis, planning, execution, verification, review, testing, and retry loops. It can coordinate multiple agent backends and persist project knowledge so later runs can reuse decisions and known fixes.
What this skill actually does
Invoke this instead of manually prompting one agent when the operator needs a supervised lifecycle for larger development work: feature implementation, debugging, refactoring, review-test-fix cycles, or UI improvement loops that may need checkpoints and adaptive retries. Upstream documents slash-command entry points such as /maestro-ralph-v2, /maestro, /maestro-quick, and Odyssey commands, plus an npm CLI installer.
The boundary is coding workflow orchestration. The skill should help an operator choose and run Maestro’s documented command chains, inspect status, resume at decision points, and keep verification explicit. It is not a general multi-agent framework listing or a replacement for the underlying coding agents.