subagent-driven-development

Dispatch isolated subagents for multi-task implementation plans with mandatory two-stage reviews.

Updated May 28, 2026
One-click install
npx skills add https://github.com/kyo0221/claude-config --skill subagent-driven-development-kyo0221
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/kyo0221/claude-config/tree/main/skills/subagent-driven-development
Command: npx skills add https://github.com/kyo0221/claude-config --skill subagent-driven-development-kyo0221

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill eliminates the inefficiency, context bloat, and quality inconsistencies of manually executing multi-task implementation plans, where constant context switching and ad-hoc reviews slow delivery and introduce avoidable errors.

Core Features & Use Cases

  • Isolated Subagent Task Execution: Dispatches a fresh, context-isolated subagent for each independent task to prevent context pollution and keep work focused.
  • Mandatory Two-Stage Review: Enforces spec compliance review first, followed by code quality review for every task to catch gaps and issues early.
  • Continuous End-to-End Execution: Runs all tasks from the implementation plan without unnecessary human check-ins, only pausing for unresolvable blockers or completed work.
  • Use Case: For a 4-task feature implementation plan, use this Skill to automatically dispatch dedicated subagents for each task, run automated reviews, and deliver a fully vetted implementation without manual coordination overhead.

Quick Start

Use the subagent-driven-development skill to execute your pre-defined implementation plan end-to-end with automated subagent task dispatch and mandatory two-stage quality reviews.

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 implementation plan execution without context pollution across tasks?▼

Automated implementation plan execution dispatches a fresh, context-isolated subagent for each independent task, preventing context pollution and keeping work focused while running tasks continuously to completion.

What is the two-stage review process for software development task automation?▼

The two-stage review process enforces spec compliance review first, followed by code quality review for every task, catching gaps and issues early before moving to the next implementation task.

How do I execute multiple independent implementation tasks in a single session?▼

Continuous end-to-end execution runs all tasks from a pre-defined implementation plan without unnecessary human check-ins, only pausing for unresolvable blockers or completed work.

Does subagent-driven workflow orchestration handle task escalations and blockers automatically?▼

Subagent workflow orchestration properly handles subagent escalations and blockers by pausing continuous execution only when encountering unresolvable blockers that require human intervention.

When do I need automated subagent dispatch for code review and task automation?▼

Automated subagent dispatch is needed for software engineering workflows with pre-defined, independent implementation tasks requiring same-session execution with strict quality gates to eliminate manual coordination overhead.

What are the limitations of using subagent workflows for implementation plan execution?▼

Subagent workflows require pre-defined, independent implementation tasks and strict quality gates, meaning poorly defined task boundaries or highly interdependent tasks may not suit this automated dispatch approach.