maestro-task

Execute single PR-scoped tasks with claim/block handoffs and verify-driven iteration.

1|Updated May 26, 2026

Search Tags:#verification #worktree #task orchestration #handoffs #maestro #PR workflow #product-spec

Ready to run

npx skills add https://github.com/Leonardo-Corte/local-social-smm-agent --skill maestro-task

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: maestro-task
Source: https://github.com/Leonardo-Corte/local-social-smm-agent/tree/main/reference/github-agent-orchestration/maestro/skills/bundled/maestro-task
Command: npx skills add https://github.com/Leonardo-Corte/local-social-smm-agent --skill maestro-task

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill prevents multi-step work from becoming chaotic by enforcing a disciplined single-task loop (spec → task → claim → verify/block → ship) with clear handoffs.

Core Features & Use Cases

Single-task product loop: Materialize exactly one PR-scoped task from a vetted spec, then iterate with verification until PASS.
Handoff-ready transitions: claim and block automatically emit handoff envelopes under .maestro/handoffs/, enabling the next agent to continue safely.
Guardrailed failure handling: Uses explicit verdicts (PASS/FAIL/HUMAN/BLOCK) with required reasons, and routes heavy-mode work to maestro-mission.

Quick Start

Use the maestro task loop in a maestro project by creating a light spec, materializing a task from it, then claiming it with an agent id and running verify until it passes.

Frequently Asked Questions about maestro-task

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I manage task handoffs for multi-step PR workflows without losing context?▼

Task handoffs in multi-step PR workflows are managed by claiming and blocking tasks, which automatically emit handoff envelopes under `.maestro/handoffs/` for the next agent. This ensures safe task transitions without losing context.

What is single-task orchestration for PR-scoped work?▼

Single-task orchestration for PR-scoped work is a disciplined loop: spec creation, materializing one task, claiming it, verifying, and shipping. It prevents chaotic multi-step execution by enforcing claim, block, and verify-driven iteration until PASS.

How do I handle FAIL verdicts during task verification?▼

Handling FAIL verdicts during task verification involves entering a repair loop that requires explicit reasons for the failure. The disciplined task loop enforces verdict routing semantics to iterate until the verification achieves a PASS.

When should I route work to heavy-mode mission orchestration instead of single-task execution?▼

Work should route to heavy-mode mission orchestration when tasks exceed single PR-scoped execution limits. Verdict routing semantics automatically direct complex, multi-step implementation requests to maestro-mission instead of single-task execution.

Do I need a maestro-initialized project to use single-task execution loops?▼

Yes, a maestro-initialized project is required to use single-task execution loops. The task lifecycle depends on maestro verbs for spec creation, task materialization, and enforcing verification semantics within that specific environment.