subagent-driven-development

Dispatch AI subagents for each implementation plan task with two-stage reviews.

3|Updated Feb 21, 2026

Search Tags:#automation #development #code review #ai agents #iterative development #task execution

Ready to run

npx skills add https://github.com/lucasgrow/some-skills --skill subagent-driven-development-lucasgrow

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/lucasgrow/some-skills/tree/main/subagent-driven-development
Command: npx skills add https://github.com/lucasgrow/some-skills --skill subagent-driven-development-lucasgrow

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill streamlines the development process by breaking down implementation plans into independent tasks, each handled by a dedicated AI subagent with a rigorous two-stage review process, ensuring high quality and fast iteration.

Core Features & Use Cases

Task Decomposition: Automatically breaks down a plan into discrete, manageable tasks.
Independent Subagents: Assigns a fresh AI subagent to each task, preventing context pollution.
Two-Stage Review: Implements a spec compliance review followed by a code quality review for each task.
Use Case: Developing a new feature with a detailed implementation plan. This Skill ensures each part of the feature is built correctly, tested, and reviewed before moving to the next, leading to a robust final product.

Quick Start

Use the subagent-driven-development skill to execute the implementation plan located at ./my-feature-plan.md.

Frequently Asked Questions about subagent-driven-development

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

What is automated iterative development using AI subagents?▼

AI subagents for iterative development work by breaking a feature plan into discrete tasks, dispatching a fresh subagent for each task, and enforcing a strict two-stage review process. This prevents context pollution and validates spec compliance and code quality before proceeding to the next task.

How do I execute an implementation plan using AI agents for task decomposition?▼

To execute an implementation plan using AI agents, you provide the plan file, such as ./my-feature-plan.md. The system automatically decomposes the plan into manageable tasks, assigns an independent AI subagent to each task, and enforces spec compliance and code quality reviews sequentially within the same session.

Does subagent-driven development prevent context pollution across multiple tasks?▼

Subagent-driven development prevents context pollution by assigning a fresh, independent AI subagent to each task within the implementation plan. This isolated execution ensures that context from one task does not interfere with subsequent development work.

What is the best way to automate code review and spec compliance for iterative development?▼

The best way to automate code review and spec compliance is to enforce a strict two-stage review process after each task's completion. The first stage checks spec compliance, and the second stage evaluates code quality, ensuring a robust final product through focused agentic execution.

Can I use this subagent-driven development methodology for a detailed feature implementation plan?▼

Yes, this methodology is designed for detailed feature implementation plans. It ensures each part of the feature is built correctly, tested, and reviewed sequentially via dedicated subagents and a two-stage review process, leading to a robust final product.

What are the limitations of using independent AI subagents for software development?▼

Using independent AI subagents for software development requires a detailed implementation plan as input and operates within a single session for the two-stage review process. It relies on the quality of the initial plan decomposition and may not handle tasks requiring cross-task state maintenance.