subagent-driven-development

Execute implementation plans by dispatching fresh subagents per task with two-stage reviews.

Updated May 13, 2026
One-click install
npx skills add https://github.com/kimprobably/maestro-os --skill subagent-driven-development-kimprobably
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/kimprobably/maestro-os/tree/main/hermes/distribution/maestro-operator/skills/subagent-driven-development
Command: npx skills add https://github.com/kimprobably/maestro-os --skill subagent-driven-development-kimprobably

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It solves the problem of delivering plan-based implementations efficiently without losing quality by orchestrating work across isolated subagent runs.

Core Features & Use Cases

  • Fresh subagent per task: Prevents subagents from inheriting prior session context so each task stays focused.
  • Two-stage review after every task: Runs a spec compliance review first, then a code quality review to catch both omissions and poor implementation.
  • Continuous execution: Keeps execution moving through all tasks in the plan without waiting for check-ins, except when blocked or complete.

Use case: Executing a multi-step engineering plan where tasks are mostly independent (e.g., implementing multiple features and tests in sequence) while maintaining strict adherence to the provided requirements.

Quick Start

Use subagent-driven-development to execute an implementation plan by dispatching an implementer subagent for each task, then running spec compliance and code quality reviews after each task before moving to the next.

Frequently Asked Questions about subagent-driven-development

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

FAQPage Schema
How do I execute an implementation plan with subagents while maintaining spec compliance?

Executing an implementation plan with subagents involves dispatching a fresh subagent per task and running a two-stage review process that checks spec compliance first, then code quality, before moving to the next task.

What is two-stage code review in subagent-driven development?

Two-stage code review in subagent-driven development runs a spec compliance review first to catch omissions, followed by a code quality review to catch poor implementation, ensuring strict engineering standards after every task.

How do I handle blocked tasks or missing context during subagent task orchestration?

Handling blocked tasks or missing context during subagent task orchestration requires tracking structured outcomes like BLOCKED or NEEDS_CONTEXT, pausing continuous execution until the specific blocker is resolved or the missing context is provided.

Can subagents run implementation tasks continuously without waiting for check-ins?

Subagents can run implementation tasks continuously without waiting for check-ins, keeping execution moving through all tasks in the plan automatically until a task is blocked or the entire plan is complete.

When should I use a fresh subagent for each task in test driven development?

You should use a fresh subagent for each task in test driven development when tasks are mostly independent but require strict spec adherence, preventing subagents from inheriting prior session context so each task stays focused.