subagent-driven-development

Dispatch fresh subagents per task to execute implementation plans with two-stage reviews.

6|Updated Feb 4, 2026
One-click install
npx skills add https://github.com/futuregerald/futuregerald-claude-plugin --skill subagent-driven-development-futuregerald
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/futuregerald/futuregerald-claude-plugin/tree/main/skills/subagent-driven-development
Command: npx skills add https://github.com/futuregerald/futuregerald-claude-plugin --skill subagent-driven-development-futuregerald

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Executing implementation plans can become error-prone in a single session without proper task isolation and review; Subagent-Driven Development introduces a two-stage review workflow using a fresh subagent per task to improve quality and speed.

Core Features & Use Cases

  • Fresh subagent per task to prevent context pollution and enable parallel/task isolation.
  • Two-stage reviews: spec compliance review first, then code quality review, with iterative fixes.
  • Plan-driven execution in a single session, supporting sequential or parallel task execution.

Quick Start

Dispatch an implementer subagent for each task in your plan, then run the two-stage reviews to ensure spec compliance and code quality.

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 without context pollution in a single session?▼

Executing implementation plans without context pollution is achieved by dispatching a fresh subagent per task, isolating each task to prevent cross-contamination and enable rapid iteration within a single session.

What is two-stage code review and how does it improve spec compliance?▼

Two-stage code review is a workflow that gates quality by running a spec compliance review first, then a code quality review, coordinating separate reviewer prompts to iteratively fix issues and ensure implementation matches the plan.

How do I dispatch independent tasks to subagents for rapid iteration?▼

You dispatch independent tasks to subagents by using a task-dispatch mechanism that extracts tasks from a plan and assigns each to a fresh implementer subagent, supporting sequential or parallel execution within the current session.

When do I need subagent task isolation for my development workflow?▼

You need subagent task isolation when your development workflow involves executing implementation plans with multiple independent tasks, requiring context pollution prevention and two-stage spec and code quality reviews to maintain quality.

Does subagent-driven development work for sequential and parallel task execution?▼

Subagent-driven development works for both sequential and parallel task execution, coordinating implementer, spec-reviewer, and code-quality-reviewer prompts to gate quality and extract tasks from plans within a single session.

Why does my single session implementation plan become error-prone without task isolation?▼

Single session implementation plans become error-prone without task isolation due to context pollution; dispatching a fresh subagent per task prevents this by isolating work and applying two-stage reviews for spec compliance and code quality.