Does it support Test-Driven Development?

Yes, subagents dispatched by this skill naturally follow TDD patterns, implementing, testing, and self-reviewing before handing back to the controller.

How does this differ from standard plan execution?

It uses fresh subagents for every task to avoid context pollution and enforces a mandatory two-step review process (spec then quality) within the same session for faster iteration.

Why is the review process split into two stages?

Splitting reviews ensures the code meets all functional requirements (Spec Compliance) before the AI spends resources refining the style or architecture (Code Quality).

Does this skill work for tightly coupled tasks?

This skill is optimized for independent tasks. If tasks are tightly coupled, manual execution or a brainstorming-first approach is recommended to manage dependencies.

Subagent-Driven Development

Name: Subagent-Driven Development
Author: abudhahir

byabudhahir

0•

Productividad y Flujo de Trabajo

Executes complex development plans by dispatching fresh subagents for individual tasks with mandatory two-stage quality reviews.

Subagent-Driven Development is a high-efficiency workflow for Claude Code designed to execute implementation plans by decomposing them into independent tasks handled by isolated subagents. This methodology prevents context pollution and ensures high-quality output by requiring each subagent to follow Test-Driven Development (TDD) principles. The skill enforces a rigorous two-stage review process—verifying specification compliance first, followed by a code quality audit—ensuring that every task meets functional requirements and engineering standards before completion, all within a single development session.

Características Principales

01Real-time task tracking via TodoWrite integration

020 GitHub stars

03Isolated subagent dispatching per task to prevent context drift

04Built-in Test-Driven Development (TDD) enforcement

05Two-stage automated review (Spec Compliance then Code Quality)

06Interactive clarification loop for subagent questions

Casos de Uso

01Executing multi-step feature roadmaps within a single session

02Implementing complex backend logic with strict specification requirements

03Large-scale refactoring requiring high-precision quality gates

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add abudhahir/superpowers subagent-driven-development

For use in Claude.ai and ChatGPT

Download Skill