subagent-driven-development

Dispatch subagents for each task with two-stage spec and code quality reviews.

16|1|Updated Dec 23, 2025
One-click install
npx skills add https://github.com/AndurilCode/mcp-apps-kit --skill subagent-driven-development-andurilcode
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/AndurilCode/mcp-apps-kit/tree/main/.claude/skills/subagent-driven-development
Command: npx skills add https://github.com/AndurilCode/mcp-apps-kit --skill subagent-driven-development-andurilcode

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Subagent-Driven Development enables teams to execute implementation plans by dispatching independent subagents for each task, with two-stage reviews after every task to ensure spec compliance and high code quality.

Core Features & Use Cases

  • Fresh subagent per task with a two-stage review process (spec compliance first, then code quality).
  • Per-task extraction of plan text and automatic dispatch to specialized subagents (implementer, spec reviewer, code quality reviewer).
  • Faster iteration within a single session by reducing handoffs and context switching.
  • Use Case: A product feature plan is broken into five tasks; for each task, an implementer subagent executes, a spec reviewer validates requirements, and a code-quality reviewer verifies tests and maintainability.

Quick Start

To begin, load your plan file, dispatch the implementer subagent with the full task text, and follow the two-stage review loops until all tasks are complete.

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 plan execution using subagents for feature implementation?

Automate plan execution by dispatching a dedicated subagent for each task within a single session. The system applies a two-stage review cycle, checking spec compliance and code quality, to ensure accurate feature implementation.

What is subagent-driven development and how does it handle code review?

Subagent-driven development is a workflow that executes implementation plans by dispatching independent subagents per task. It handles code review through a two-stage process: a spec reviewer validates requirements, followed by a code-quality reviewer verifying tests and maintainability.

How do I start a subagent-driven workflow for a software development plan?

To start subagent-driven development, load your plan file and dispatch the implementer subagent with the full task text. The system automatically extracts plan text per task and initiates the two-stage review loops until all tasks are complete.

Can I use subagent-driven development to reduce context switching during iterative development?

Yes, subagent-driven development enables faster iteration within a single session by reducing handoffs and context switching. It achieves this by using a fresh subagent per task with standardized prompt templates for implementation and review.

Does subagent-driven development require external dependencies to dispatch tasks?

No, subagent-driven development operates without external dependencies. It uses standardized prompt templates to dispatch specialized subagents for implementation, spec compliance, and code quality verification directly within the workflow.

Why does subagent-driven development use a two-stage review cycle for task execution?

The two-stage review cycle ensures spec compliance and high code quality by separating concerns. A spec reviewer first validates that requirements are met, then a code-quality reviewer verifies tests and maintainability before proceeding to the next task.