Do I need a plan before using this skill?

Yes, this skill is specifically designed to execute an existing implementation plan (created via skills like superpowers:writing-plans) where tasks are relatively independent.

What is the benefit of using a fresh subagent for every task?

Using a fresh subagent prevents 'context pollution,' ensuring the AI doesn't get confused by previous task details and remains strictly focused on the current requirement and codebase state.

Can the implementer subagent ask me questions during the process?

Absolutely. The workflow encourages the subagent to surface questions for clarification before and during implementation to ensure the final output meets your expectations.

How does the two-stage review process work?

First, a reviewer subagent checks if the code matches the specific requirements (spec compliance). Once passed, a second review assesses code quality, naming conventions, and architectural best practices.

Subagent-Driven Development

Name: Subagent-Driven Development
Author: dkyazzentwatwa

bydkyazzentwatwa

•

생산성 및 워크플로

Executes complex implementation plans by dispatching specialized AI subagents for task execution and multi-stage code reviews.

소개

Subagent-Driven Development is a specialized workflow skill for Claude Code designed to execute multi-task development plans with high precision and minimal context pollution. By dispatching a fresh subagent for every independent task within the same session, it ensures that implementation remains focused and clean. The skill enforces a rigorous two-stage quality gate: a specification compliance review followed by a code quality review. This structured approach allows for rapid iteration, automated test-driven development, and superior code quality without the overhead of manual context switching or parallel session management.

주요 기능

Automated TDD-based implementation cycles for every task
29 GitHub stars
Integrated TodoWrite tracking for real-time progress monitoring
Zero-handoff same-session execution for faster development cycles
Fresh subagent dispatching per task to eliminate context pollution
Two-stage review protocol: Specification Compliance then Code Quality

사용 사례

Maintaining high code quality in complex codebases through automated review loops
Scaling development velocity by automating the implementation and testing of plan items
Executing feature plans with multiple modular or independent sub-tasks

소개

주요 기능

Automated TDD-based implementation cycles for every task
29 GitHub stars
Integrated TodoWrite tracking for real-time progress monitoring
Zero-handoff same-session execution for faster development cycles
Fresh subagent dispatching per task to eliminate context pollution
Two-stage review protocol: Specification Compliance then Code Quality

사용 사례

Maintaining high code quality in complex codebases through automated review loops
Scaling development velocity by automating the implementation and testing of plan items
Executing feature plans with multiple modular or independent sub-tasks