subagent-driven-development

Dispatch fresh subagents per task with automated testing and code reviews.

Updated Oct 29, 2025
One-click install
npx skills add https://github.com/hmps/skills --skill subagent-driven-development-hmps
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/hmps/skills/tree/main/subagent-driven-development
Command: npx skills add https://github.com/hmps/skills --skill subagent-driven-development-hmps

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill automates the execution of complex development plans by dispatching specialized AI subagents for each task, with integrated code reviews. It ensures high-quality, fast iteration by catching issues early, eliminating context switching, and maintaining a pristine codebase.

Core Features & Use Cases

  • Task-Specific Subagents: Dispatches a fresh subagent for each task, preventing context pollution and ensuring focused, efficient execution.
  • Integrated Code Review: Automatically triggers a code review after each task, providing continuous quality gates and immediate feedback.
  • Use Case: After generating an implementation plan with writing-plans, use this Skill to have AI subagents systematically implement each task, review their own work, and report back, accelerating development while maintaining rigorous quality standards.

Quick Start

Use the subagent-driven-development skill to execute the implementation plan for the new user dashboard. Start with Task 1: "Set up database schema."

Frequently Asked Questions about subagent-driven-development

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

FAQPage Schema
How do I automate code review in my development workflow?

Automate code review by dispatching AI subagents to execute tasks and trigger mandatory review gates after each one. This catches issues early, eliminates context switching, and maintains code quality without manual bottlenecks.

What's the best way to structure a complex development plan with multiple tasks?

Structure plans by breaking them into independent, task-specific units. Assign fresh subagents to each task to prevent context pollution, enable focused execution, and allow rapid iteration with integrated quality gates between tasks.

Can I use AI subagents to execute implementation tasks in isolation?

Yes. Dispatch a fresh subagent per task to ensure isolated execution, prevent cross-task interference, and enable fast feedback loops. Each subagent operates independently and reports back after completing its unit of work.

How do subagent-driven development and code review integration work together?

Subagent-driven development pairs task-specific agent execution with automated code review gates. After each subagent completes its task, a review step validates quality before the next task begins, ensuring rigorous standards throughout.

When should I use per-task subagent dispatch instead of a single agent?

Use per-task dispatch when your plan contains mostly independent tasks requiring isolation, fast feedback, and mandatory reviews between stages. This approach prevents context drift and maintains pristine codebase integrity.

What's required before running an implementation plan with subagents?

Generate a structured implementation plan first using planning tools. Break it into clear, independent tasks with defined acceptance criteria. Then dispatch subagents per task, enabling each to execute and report back with code review validation.