Skip to content
maestro
Esc
  • IntroductionWhat OMP Maestro is, what it installs, and which parts are Maestro's own code.
  • InstallationInstall Maestro with npx or bunx, pick skill libraries, log in to providers and give the router a key.
  • EnsemblesThe three model ensembles — mix, anthropic and openai — and every role they assign.
  • Jev RouterHow Maestro's router uses TypeSafe Jev for main-turn downgrades and subagent worker classes, and how it fails open.
  • SkillsThe skill libraries Maestro can install, who made them, and how to add your own.
  • CommandsEvery command Maestro adds, plus the omp and ponytail commands it leans on.
  • ConfigurationCustomise roles, fallback chains, ensembles and skill libraries — and what Maestro does and doesn't touch.
  • Troubleshooting & FAQFixes for common setup problems and answers to frequent questions.

Introduction

What OMP Maestro is, what it installs, and which parts are Maestro's own code.

OMP Maestro is an opinionated setup for the oh-my-pi (omp) coding-agent harness. One command installs a curated set of community plugins and skills, writes a model ensemble into your omp config, and adds Maestro's own Jev Router extension.

Terminal
$ npx omp-maestro

What Maestro is

omp already gives you model roles, fallback chains, subagents, plugins and skills. Maestro is one specific, reproducible way of filling them in — the setup its author uses every day, packaged so you can try it in one go and change anything you don't like.

  • Three ensembles — mix, anthropic and openai — each a complete set of model roles and fallback chains, switchable with /maestro.
  • The Jev Router — an omp extension that asks TypeSafe Jev how much model each piece of work needs, and fails open to your normal roles when it can't answer.
  • Curated skills — 6 libraries, up to 52 skills, installed from their authors' repositories through an arrow-key checklist.

What it is not

  • Not a model. Every model it configures comes from Anthropic or OpenAI; the router's judge is TypeSafe Jev.
  • Not a replacement for omp. Maestro is an omp plugin plus an installer; omp does all the actual work.
  • Not an original plugin ecosystem. Most of what gets installed is other people's work, installed from the original sources and credited by name.

What gets installed

PieceSourceWhat it adds
omp-maestrohsmoviich/omp-maestroJev Router, /maestro, the smart-commit agent
ponytailDietrichGebert/ponytail“Lazy senior dev” mode — the smallest complete change
up to 52 skillsskills.json, via the skills CLIThe libraries you tick in the checklist
Model roles + fallbacksprofiles.jsonWritten to your omp config — only the roles an ensemble lists

Who wrote what

Being precise about this is the point of the project, so here it is in one place:

  • Maestro (MIT): the Jev Router, the /maestro command, the installer, the smart-commit agent definition and the three ensembles.
  • omp (native): model roles, retry.fallbackChains, the task subagent tool, /vibe, /model, /login, plugins and skills loading.
  • TypeSafe: the Jev judge model and the @typesafe-ai/sdk client the router calls.
  • Community: ponytail, the skills CLI and every skill library — each keeps its own license.

Version

These docs are generated against omp-maestro@0.1.2. Ensembles, skill lists and router thresholds are read from the repository at build time.

Next steps