subagent-driven-development

Dispatches a fresh subagent per implementation task with two-stage spec and code quality reviews.

Updated Apr 3, 2026
One-click install
npx skills add https://github.com/norrbacka/threejs_codex_base --skill subagent-driven-development-norrbacka
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/norrbacka/threejs_codex_base/tree/main/.codex/superpowers/skills/subagent-driven-development
Command: npx skills add https://github.com/norrbacka/threejs_codex_base --skill subagent-driven-development-norrbacka

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you execute implementation plans without losing focus by delegating each independent task to a fresh subagent and reviewing results before moving on.

Core Features & Use Cases

  • Fresh subagent per task to keep each task isolated from the rest of the session.
  • Two-stage review flow with spec compliance review first and code quality review second.
  • Plan execution support for same-session work such as feature slices, bug fixes, and coordinated refactors.
  • Task tracking and handoff discipline so you can finish one task cleanly before starting the next.

Quick Start

Use this skill to execute your current implementation plan task by task with a fresh subagent and both review stages after each task.

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 implementation plans without context pollution in coding agents?

You execute implementation plans without context pollution by dispatching a fresh subagent for each independent task. This isolates task execution, preventing the accumulated context of previous tasks from degrading the agent's focus and output quality.

What is a two-stage code review workflow for subagent task orchestration?

A two-stage code review workflow for task orchestration applies spec compliance first and code quality second. This ensures the generated code meets the original plan requirements before evaluating structural quality, maintaining high standards in subagent-driven development.

When do I need plan-driven task orchestration for development work?

You need plan-driven task orchestration for same-session development work where tasks are mostly independent. It is ideal for executing feature slices, bug fixes, and coordinated refactors that require isolated execution and clean task handoffs.

How to manage task tracking and branch finishing in subagent-driven development?

You manage task tracking and branch finishing by completing one task cleanly before starting the next. After passing both review stages, the workflow applies final branch finishing to ensure the coordinated execution results in a coherent codebase.

Does subagent-driven development work for coordinated refactors in the same session?

Yes, subagent-driven development works for coordinated refactors in the same session. By delegating each mostly independent task to a fresh subagent, it supports coordinated execution while maintaining task isolation and tracking discipline.