subagent-driven-development

Dispatch fresh subagents per task and run two-stage reviews.

Updated Jan 28, 2026
One-click install
npx skills add https://github.com/kckylechen1/ai-agent-toolkit --skill subagent-driven-development-kckylechen1
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/kckylechen1/ai-agent-toolkit/tree/main/skills/subagent-driven-development
Command: npx skills add https://github.com/kckylechen1/ai-agent-toolkit --skill subagent-driven-development-kckylechen1

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill orchestrates per-task execution by dispatching dedicated subagents within a single session, enabling rapid iteration without cross-task context contamination.

Core Features & Use Cases

  • Fresh subagent per task to maintain clean context and improve focus.
  • Two-stage review workflow: spec compliance review followed by code quality review.
  • Integrated templates and prompts to automate task dispatch and review loops for iterative development.

Quick Start

Provide your implementation plan and trigger per-task subagents, then run the two-stage reviews for each task and iterate until all tasks are approved.

Frequently Asked Questions about subagent-driven-development

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

FAQPage Schema
How do I automate task dispatch and prevent context contamination across implementation tasks?

Spec compliance review verifies that the subagent's output meets the original requirements, while code quality review checks the implementation standards. This two-stage workflow ensures high quality before finalizing any task.

How do I execute an implementation plan step by step using subagents?

To execute an implementation plan, provide the plan to trigger per-task subagents sequentially. Run the two-stage review process for each task and iterate until all tasks are approved and finalized.

Do I need to provide my own implementation plan before using subagent-driven development?

Yes, you need to provide your own implementation plan as input. The skill uses per-task prompts and templates to orchestrate the subagents based on the specific tasks outlined in your plan.

What is the best way to ensure spec compliance during rapid code iteration?

The best way to ensure spec compliance is a two-stage review workflow that checks spec compliance first, followed by code quality. This structured approach maintains standards during fast iteration.

Why use a fresh subagent for each task instead of a single agent for an entire implementation plan?

Using a fresh subagent per task maintains clean context and improves focus. It prevents the accumulated context from previous tasks from degrading performance or causing cross-task contamination during rapid iteration.