subagent-driven-development

Dispatch implementation tasks to AI subagents with two-stage review.

1|Updated May 25, 2025

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

Ready to run

npx skills add https://github.com/circld/dotfiles --skill subagent-driven-development-circld

💬 Prompt for ChatGPT / Claude Chat▼

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

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 fresh AI subagent with a two-stage review process, ensuring high quality and rapid iteration within a single session.

Core Features & Use Cases

Task Decomposition: Executes complex plans by dispatching individual tasks to specialized subagents.
Two-Stage Review: Implements a rigorous review process: first for spec compliance, then for code quality.
Use Case: When you have a detailed feature plan with multiple distinct steps, use this Skill to have AI agents implement each step sequentially, with automated checks at each stage, before moving to the next.

Quick Start

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

Frequently Asked Questions about subagent-driven-development

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

FAQPage Schema

How do I execute complex implementation plans using AI agents?▼

You can execute complex implementation plans by dispatching independent tasks to specialized AI subagents within a single session. The Skill breaks down plans into discrete steps, managing sequential execution for rapid iterative development.

How does the two-stage review process work for AI-generated code?▼

The two-stage review process first checks spec compliance to ensure the task meets requirements, then evaluates code quality. This automated checking occurs after each subagent completes its task before moving to the next step.

What is the best way to manage iterative development with AI agents without losing context?▼

The best way to manage iterative development is by using subagents to work on discrete, well-defined tasks. This approach minimizes context switching by orchestrating execution within a single session while maintaining high-quality output.

Do I need a detailed feature plan before using subagent-driven development?▼

Yes, you need a detailed feature plan with multiple distinct steps. The Skill executes the plan located at a specified path, dispatching each individual step to a fresh AI subagent for implementation.

Can I use this workflow for tasks that require sequential code review and quality checks?▼

Yes, this workflow is designed for sequential tasks requiring code review. Each subagent's work undergoes a rigorous two-stage review for spec compliance and code quality, ensuring high standards before proceeding to the next task.