When is the best time to use this skill?

It is best used when you have a detailed implementation plan with independent tasks and wish to remain in your current session for faster iteration without the overhead of context switching.

What happens if a subagent fails a quality or spec review?

The workflow mandates a fix-and-review loop. The implementer subagent must address the specific issues identified by the reviewer and pass both review stages before the task is marked complete.

How does Subagent-Driven Development differ from standard plan execution?

Unlike standard execution which might handle tasks in bulk, this skill dispatches a fresh subagent for every task to prevent context pollution and enforces a strict two-stage review cycle (spec compliance then code quality) for every unit of work.

Does this skill support parallel task execution?

No. To avoid git conflicts and maintain a clean session state, tasks are executed sequentially, but the automated review loops make this process significantly faster than manual execution.

Subagent-Driven Development

Name: Subagent-Driven Development
Author: claudiodearaujo

byclaudiodearaujo

0•

Productivity & Workflow

Executes complex implementation plans by orchestrating specialized subagents for task implementation and multi-stage quality reviews.

Subagent-Driven Development is a specialized workflow skill designed to automate the execution of multi-task development plans within a single Claude session. It functions by dispatching a fresh implementer subagent for each independent task, followed by a rigorous two-stage review process: first verifying specification compliance and then enforcing high code quality. This approach minimizes context pollution, ensures adherence to Test-Driven Development (TDD) principles, and provides a continuous progress loop that brings features from plan to production-ready code with minimal human intervention.

Key Features

01Proactive subagent questioning to resolve ambiguities before implementation starts.

02Automatic task tracking and status management using TodoWrite integration.

03Mandatory two-stage review process covering both spec compliance and code quality.

040 GitHub stars

05Integrated Test-Driven Development (TDD) requirements for every subagent task.

06Fresh subagent dispatching per task to eliminate context drift and confusion.

Use Cases

01Executing a multi-step feature roadmap where tasks are mostly independent.

02Rapid prototyping of complex systems requiring high-fidelity code and full test coverage.

03Systematic refactoring of large modules with automated verification at every step.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add claudiodearaujo/izacenter subagent-driven-development

For use in Claude.ai and ChatGPT

Download Skill