subagent-driven-development

Execute implementation plans by delegating tasks to fresh subagents with staged reviews.

Updated Apr 2, 2026
One-click install
npx skills add https://github.com/JKhyro/HERMES-AGENT --skill subagent-driven-development-jkhyro
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/JKhyro/HERMES-AGENT/tree/main/skills/software-development/subagent-driven-development
Command: npx skills add https://github.com/JKhyro/HERMES-AGENT --skill subagent-driven-development-jkhyro

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill removes the risk and slowdown of implementing a plan in one long, error-prone session by splitting work into fresh subagent tasks with structured review gates.

Core Features & Use Cases

  • Plan-to-task execution: Reads an implementation plan once, extracts every task, and turns the plan into a clear task list.
  • Fresh subagent per task: Dispatches a new implementer for each task so context stays clean and work stays focused.
  • Two-stage review: Verifies each task first for spec compliance and then for code quality before moving on.
  • Best for: Feature builds, refactors, multi-file changes, and any workflow where correctness and consistency matter across many steps.

Quick Start

Use this skill to execute your implementation plan task by task with a fresh subagent for each item and two review passes before completion.

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 a multi-step implementation plan without losing context across files?

Subagent-driven development prevents context loss by dispatching a fresh subagent for each extracted task in your implementation plan. This keeps work focused and context clean before final integration validation.

How does two-stage code review work for multi-file refactoring?

Two-stage code review verifies each completed task first for spec compliance and then for code quality. This staged review gate ensures correctness before moving on to the next task.

Can I use this workflow for complex feature builds and multi-file refactors?

Yes, this workflow is best for feature builds, refactors, and multi-file changes where correctness and consistency matter across many steps. It applies structured execution gates to deliver features safely.

What's the best way to coordinate independent implementation tasks in software engineering?

The best way is delegating each independent task to a fresh subagent with structured review gates. This requires full task extraction, spec compliance checks, code quality review, and final integration validation before completion.

When should I not use subagent delegation for implementation plans?

You should not use subagent delegation for single-step changes or workflows lacking a structured implementation plan. This approach requires multi-step software changes and full task extraction to coordinate staged reviews effectively.