Is this skill suitable for highly coupled tasks?

It is best used for tasks that are mostly independent. If tasks are extremely tightly coupled or the plan requires frequent revision, manual execution or brainstorming may be more effective.

What is the main benefit of using subagents over manual execution?

Subagents provide fresh, isolated context for every task. This prevents context pollution and hallucinations, while ensuring that Test-Driven Development patterns are followed consistently.

Do I need a pre-existing plan to use this skill?

Yes, this skill is designed to execute plans created by the 'writing-plans' skill. It reads the plan file and converts it into a structured Todo list for execution.

How does the automated code review process work?

After a subagent completes a task, a separate code-review-agent is dispatched to evaluate the work against requirements. It identifies critical or minor issues that must be fixed before the next task begins.

Does using subagents increase the cost of development?

While it involves more subagent invocations, it reduces overall cost and time by catching errors early through automated reviews, preventing expensive debugging sessions later in the project.

Subagent-Driven Development

Name: Subagent-Driven Development
Author: cipherstash

bycipherstash

•

生産性とワークフロー

Executes complex development plans by dispatching specialized subagents for per-task implementation followed by automated code reviews.

Subagent-Driven Development is a sophisticated workflow skill for Claude Code that automates the execution of multi-step implementation plans. By spawning a fresh, context-isolated subagent for every individual task and enforcing an automated code review cycle between steps, it ensures high-quality code output while preventing the context pollution that often occurs in long-running AI sessions. This skill seamlessly integrates task tracking, Test-Driven Development (TDD), and branch finalization, making it the ideal choice for developers looking to scale their output with consistent quality gates and minimal manual intervention.

主な機能

01Integrated TodoWrite task tracking for real-time progress monitoring

02Automated code review cycles between every implementation step

034 GitHub stars

04Seamless integration with Test-Driven Development (TDD) practices

05Automated branch finalization and requirement verification

06Context-isolated subagent dispatch per task to prevent halluncinations

ユースケース

01Executing multi-feature development plans within a single session

02Refactoring large codebases where tasks are modular and independent

03Scaling coding output through automated quality gates and review loops

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add cipherstash/cipherpowers subagent-driven-development

For use in Claude.ai and ChatGPT

Download Skill