subagent-driven-development

Dispatch independent subagents to execute plan tasks with isolated context.

1|Updated Mar 8, 2024
One-click install
npx skills add https://github.com/cpp-for-everything/ORM --skill subagent-driven-development-cpp-for-everything
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/cpp-for-everything/ORM/tree/main/.windsurf/skills/subagent-driven-development
Command: npx skills add https://github.com/cpp-for-everything/ORM --skill subagent-driven-development-cpp-for-everything

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The Subagent-Driven Development Skill enables executing implementation plans by dispatching independent subagents per task, with isolated context to maintain focus and reduce cross-task contamination.

Core Features & Use Cases

  • Fresh subagent per task to avoid context leakage
  • Two-stage review after each task: spec compliance, then code quality
  • Clear handoffs and templates to coordinate multi-step workflows
  • Ideal for plans with largely independent tasks requiring rapid iteration

Quick Start

Dispatch the first implementer subagent to start Task 1 from your plan and trigger the two-stage review workflow after each task.

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 independent coding tasks without cross-task context leakage?

Dispatching fresh subagents per task isolates context to prevent cross-task contamination. This approach enables focused execution and rapid iteration for largely independent implementation plan tasks.

How does a two-stage code review workflow validate spec compliance and code quality?

A two-stage code review workflow evaluates spec compliance first, then code quality, after each task. Structured handoffs and prompt templates coordinate these staged reviews to maintain implementation standards.

What is the best way to dispatch subagents for rapid task iteration?

Dispatching independent subagents to execute plan tasks provides rapid iteration. Using isolated context and structured prompt templates for handoffs ensures focused task execution and strict staged reviews.

Can I use subagent-driven development for TDD tasks with isolated contexts?

Subagent-driven development supports TDD tasks by using isolated context for each subagent. Fresh subagents execute independent tasks while structured templates coordinate handoffs and trigger two-stage reviews.

When should I avoid dispatching isolated subagents for task execution?

Avoid dispatching isolated subagents for tightly coupled tasks with heavy dependencies. This approach satisfies plans where tasks are largely independent and require strict, staged reviews after each task.