How does this differ from standard plan execution?

Unlike standard execution, SDD uses a fresh subagent for every task to prevent context drift and enforces a mandatory code review gate before moving to the next item in the plan.

How does it handle errors found during code review?

If a code reviewer subagent identifies critical issues, the skill dispatches a follow-up subagent to fix the specific problems before marking the task as complete.

What is Subagent-Driven Development?

It is an AI workflow where a primary agent manages the execution of a plan by dispatching specialized subagents to handle specific tasks, ensuring clean context and quality control.

When should I avoid using Subagent-Driven Development?

Avoid using this skill if your tasks are extremely tightly coupled or if the underlying development plan requires further brainstorming and revision before execution.

Does this skill require other plugins or skills?

Yes, it is designed to integrate with the 'writing-plans' skill for plan creation and the 'requesting-code-review' skill for the quality gate phase.

Subagent-Driven Development

Name: Subagent-Driven Development
Author: seanGSISG

byseanGSISG

생산성 및 워크플로

Automates the execution of complex development plans by dispatching isolated subagents for individual tasks with integrated code review gates.

소개

Subagent-Driven Development (SDD) is a specialized workflow skill for Claude Code designed to execute implementation plans with high precision and quality. By spawning a fresh subagent for every task, the skill prevents context pollution and ensures that each code change is focused and test-driven. The process incorporates mandatory code review checkpoints between tasks, allowing the system to identify and fix issues immediately before proceeding, resulting in faster iteration cycles and a cleaner final codebase without requiring constant human intervention.

주요 기능

Context-isolated task execution via fresh subagents
Enforced TDD patterns for implementation subagents
0 GitHub stars
Automated code review triggers after every completed task
Seamless progress tracking using TodoWrite integration
In-session execution to eliminate manual context switching

사용 사례

Executing multi-step implementation plans for new feature development
Maintaining high code quality in large sessions through automated quality gates
Scaling development workflows where tasks are independent but require verification

소개

주요 기능

Context-isolated task execution via fresh subagents
Enforced TDD patterns for implementation subagents
0 GitHub stars
Automated code review triggers after every completed task
Seamless progress tracking using TodoWrite integration
In-session execution to eliminate manual context switching

사용 사례

Executing multi-step implementation plans for new feature development
Maintaining high code quality in large sessions through automated quality gates
Scaling development workflows where tasks are independent but require verification