skills.rest
xiaoquqi
xiaoquqiCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

subagent-driven-development

Delegate independent implementation tasks to fresh subagents with automated reviews.

Updated May 1, 2026
Search Tags:#automation#workflow#planning#testing#delegation#subagent#code-review
One-click install
npx skills add https://github.com/xiaoquqi/hermes-agent-skills --skill subagent-driven-development-xiaoquqi
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/xiaoquqi/hermes-agent-skills/tree/main/software-development/subagent-driven-development
Command: npx skills add https://github.com/xiaoquqi/hermes-agent-skills --skill subagent-driven-development-xiaoquqi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Coordinating implementation plans often requires managing multiple independent tasks, ensuring fresh context per task and rigorous reviews to maintain quality and speed.

Core Features & Use Cases

  • Fresh subagent per task to avoid cross-task state.
  • Two-stage review process: spec compliance followed by code quality.
  • Scalable orchestration of plans produced by planning skills, enabling parallel task execution with traceable outcomes.

Quick Start

Dispatch a subagent for each plan task, then run spec and quality reviews before marking tasks complete.

Frequently Asked Questions about subagent-driven-development

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

FAQPage Schema
How do I orchestrate multiple independent coding tasks to prevent cross-task state contamination?▼

Orchestrating independent coding tasks requires dispatching a fresh subagent per task. This approach isolates execution context, ensuring deterministic outcomes and preventing cross-task state contamination during parallel implementation.

What is the best way to automate spec compliance and code quality reviews for an implementation plan?▼

Automating spec compliance and code quality reviews is best handled by a two-stage review process. After dispatching subagents for plan tasks, you run spec verification followed by code quality checks before marking tasks complete.

How do I scale parallel task execution for plans produced by planning skills?▼

Scaling parallel task execution involves delegating independent plan tasks to fresh subagents. This enables scalable orchestration of plans produced by planning skills, ensuring traceable outcomes and isolated execution.

Can I use subagent-driven development for workflows with critical spec compliance requirements?▼

Subagent-driven development is suitable for workflows where spec compliance and code quality are critical. It applies to plans with multiple independent tasks, ensuring automated reviews and deterministic execution.

When should I avoid delegating tasks to fresh subagents for implementation plans?▼

Delegating tasks to fresh subagents is not ideal for plans lacking multiple independent tasks. If your workflow does not require isolated context or rigorous automated reviews, this orchestration approach adds unnecessary overhead.

How does a fresh subagent per task improve workflow automation and code review?▼

A fresh subagent per task improves workflow automation by maintaining isolated context for each execution. This isolation supports a deterministic two-stage review process, ensuring high code quality and strict spec compliance.

Related Skills

Explore More Software Engineering Skills→
galangryandanagalangryandana
20

subagent-driven-development

Dispatch fresh subagents to execute independent implementation tasks with two-stage reviews.

Community
Advanced
dmedina-devdmedina-dev
1

subagent-driven-development

Execute implementation plans by dispatching fresh subagents for each task.

Community
Advanced
cheezycheezy
5

stride-subagent-workflow

Dispatch decomposition, exploration, planning, and review agents based on task metadata.

Community
Advanced
royhenengelroyhenengel

run-prompt

Execute prompts from ./prompts/ in isolated sub-task contexts.

Community
Intermediate