subagent-driven-development

Dispatch independent tasks to fresh subagents with two-stage reviews and completion tracking.

178|24|Updated Jan 14, 2026
One-click install
npx skills add https://github.com/ZhanlinCui/Ultimate-Agent-Skills-Collection --skill subagent-driven-development-zhanlincui
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/ZhanlinCui/Ultimate-Agent-Skills-Collection/tree/main/subagent-driven-development
Command: npx skills add https://github.com/ZhanlinCui/Ultimate-Agent-Skills-Collection --skill subagent-driven-development-zhanlincui

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill enables teams to execute complex implementation plans by dispatching fresh subagents to tackle independent tasks in a single session, with built-in two-stage reviews after each task to ensure spec compliance and code quality.

Core Features & Use Cases

  • Two-stage review workflow: spec compliance review followed by code quality review for every task.
  • Task dispatch and isolation: per-task subagents operate with fresh context to avoid pollution.
  • Completion tracking: automatic marking of tasks complete and seamless handoff to the next task.

Quick Start

Dispatch your plan to the system and let the first implementer subagent begin with the plan text and context.

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 implementation plans with subagents in a single session?

Dispatching independent tasks to fresh subagents enables execution of implementation plans within a single session, utilizing isolated context for each task, followed by two-stage spec and code quality reviews.

What is two-stage code review in plan-driven development?

Two-stage code review in plan-driven development first checks task implementation for spec compliance, then evaluates code quality, ensuring independent subagent tasks meet both functional and structural standards before completion tracking.

How do I automate task dispatch and completion tracking for independent development tasks?

Automating task dispatch and completion tracking uses a Todo system to hand off independent tasks to fresh subagents, automatically marking each task complete after two-stage reviews pass and seamlessly beginning the next task.

Does subagent-driven development work for single-session workflows with independent tasks?

Subagent-driven development is designed specifically for single-session workflows where tasks are largely independent, dispatching fresh subagents per task to avoid context pollution and completing automated two-stage reviews before handoff.

When should I use fresh subagents for code implementation and task isolation?

Use fresh subagents for code implementation when executing plan-driven workflows with largely independent tasks that must be completed in a single session, ensuring task isolation prevents context pollution and maintains code quality.

Why use two-stage reviews instead of a single code review pass for subagent task completion?

Two-stage reviews verify spec compliance first, then assess code quality separately, ensuring subagent implementations meet both the original plan requirements and structural standards before the Todo system marks task completion.