subagent-driven-development

Dispatch fresh subagents for each task with two-stage reviews.

321|26|Updated Mar 10, 2026
One-click install
npx skills add https://github.com/mkurman/tamux --skill subagent-driven-development-mkurman
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/mkurman/tamux/tree/main/skills/development/superpowers/subagent-driven-development
Command: npx skills add https://github.com/mkurman/tamux --skill subagent-driven-development-mkurman

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Subagent-Driven Development coordinates task execution by dispatching fresh subagents for each task, enabling focused work and rapid iteration without context bleed.

Core Features & Use Cases

  • Fresh subagent per task to keep context isolated and allow parallel work.
  • Two-stage review after each task: spec compliance review first, then code quality review.
  • Use Case: break a plan into discrete tasks and delegate each to its own subagent to accelerate delivery while maintaining quality.

Quick Start

Train your AI to initiate a subagent-driven development workflow for your current plan and start executing tasks.

Frequently Asked Questions about subagent-driven-development

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

FAQPage Schema
How do I delegate coding tasks to subagents for isolated execution?

Subagent-driven development delegates discrete tasks to fresh subagents, keeping context isolated to prevent context bleed and enable rapid iteration during in-session task execution.

What is the two-stage code review process for subagent task execution?

The two-stage review process evaluates each completed task sequentially: first checking spec compliance to verify requirements, then performing a code quality review to maintain standards before iterative rework.

How does automated task planning work with subagent-driven development?

Automated task planning breaks down your current plan into discrete tasks within a single session, then delegates each task to its own subagent using templates for implementer, spec reviewer, and code quality reviewer prompts.

Can I use subagent-driven development for parallel task orchestration?

Yes, subagent-driven development supports flexible task orchestration by dispatching fresh subagents per task, which allows parallel work and automated handoffs while maintaining modular reviews within a single session.

What are the limitations of using fresh subagents for each development task?

Subagent-driven development operates within a single session and relies on iterative rework until tasks are solved, meaning coordination is bounded by session limits and requires structured templates for each review stage.