subagent-driven-development

Orchestrate plan-driven development tasks across subagents with two-stage reviews.

2|Updated Sep 13, 2025
One-click install
npx skills add https://github.com/s977043/plangate --skill subagent-driven-development-s977043
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/s977043/plangate/tree/main/.claude/skills/subagent-driven-development
Command: npx skills add https://github.com/s977043/plangate --skill subagent-driven-development-s977043

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Delegates implementation tasks to specialized subagents with a two-stage review process to ensure spec compliance and high quality for complex plans.

Core Features & Use Cases

  • Two-stage reviews (Spec Review + Quality Review) ensure that specs are followed and implementations meet quality standards.
  • Task decomposition: automatically extracts tasks from a plan and assigns them to dedicated subagents for parallel execution.
  • Guardrails: enforces plan loading, task isolation, and fast feedback loops to accelerate development while maintaining safety.

Quick Start

Run the ai-dev-workflow plan to load the plan, then dispatch each task to a dedicated subagent for execution and subsequent reviews.

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 decomposition for parallel AI subagent execution?▼

To automate task decomposition, this skill extracts action items from a loaded plan and assigns each to a dedicated subagent for parallel execution. It enforces task isolation to maintain safety and accelerate the development workflow.

What is a two-stage review process in AI-driven development workflows?▼

A two-stage review process separates spec compliance from code quality. This skill enforces a Spec Review to ensure plans are followed, followed by a Quality Review to verify the implementation meets high standards before completion.

How do I ensure spec conformity when delegating coding tasks to AI subagents?▼

Ensuring spec conformity requires strict guardrails and automated reviews. This skill enforces plan loading and uses a dedicated spec review stage to verify that subagent outputs strictly match the original development requirements.

Do I need a predefined plan to start subagent-driven development?▼

Yes, a predefined plan is required to start subagent-driven development. The workflow mandates loading a plan first, which provides tasks for decomposition and serves as the baseline for the subsequent spec review stage.

What are the limitations of using parallel subagents for complex plan execution?▼

Without strict guardrails, parallel subagents risk losing spec conformity during complex plan execution. This skill mitigates this by enforcing task isolation and mandatory two-stage reviews, though these safety checks may slow down fast feedback loops.