# Doriandarko/maestro

A framework for Claude Opus to intelligently orchestrate subagents.

Repository: https://github.com/Doriandarko/maestro
Canonical: https://ross.abutalabs.com/products/doriandarko-maestro
Language: Python
License Family: other
Last push: 2024-07-01T06:49:15+00:00

## Health v2 (maintenance only)
Score: 25/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 64
- inputs: {"age_days": 897, "days_push": 793, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4357, forks 646 (observed 2026-08-28T04:08:46.481505+00:00)

## What it is
Maestro is a Python framework that uses an orchestrator LLM (originally Claude Opus) to break down objectives into sub-tasks, delegate them to subagent models, and refine results into a final output. It supports multiple providers including Anthropic, OpenAI, Gemini, Groq, and local models via Ollama or LM Studio through LiteLLM.

## Use cases
- break down a complex objective into sub-tasks executed by LLM subagents
- orchestrate multiple AI models to complete a goal
- run multi-agent workflows with local models via Ollama
- use Claude Opus as an orchestrator with cheaper models as workers
- add web search to subagent task creation with Tavily
- compare orchestration across OpenAI, Anthropic, Gemini, and Groq APIs

## When to choose
- you want a lightweight, hackable multi-agent orchestration script in Python
- you need an orchestrator-worker pattern with model flexibility via LiteLLM
- you want to run agent workflows locally with Ollama or LM Studio

## When to avoid
- you need a production-grade agent framework with robust error handling and observability
- you require a supported license for commercial use
- you need a polished UI or long-running agent infrastructure

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, llm-inference, prompt-engineering, rag
- domain: artificial-intelligence, large-language-models, developer-tools
- platform: python, cli, cross-platform
- tags: orchestration, subagents, anthropic, claude, litellm, ollama, task-decomposition, multi-agent, ai-agents

## Member repositories
- Doriandarko/maestro (main) score 25

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:46.481505+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:21:29.145396+00:00, confidence not recorded.
  - readme: https://github.com/Doriandarko/maestro (fetched 2026-08-28T04:08:46.481505+00:00, sha 192d57a515ad)
- Data as of 2026-08-30T08:39:29.467469+00:00.