subagent-driven-development

Dispatch subagents to implement plan tasks with mandatory review gates.

Updated Feb 28, 2026
One-click install
npx skills add https://github.com/phant-app/phant --skill subagent-driven-development-phant-app
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/phant-app/phant/tree/main/.agents/skills/subagent-driven-development
Command: npx skills add https://github.com/phant-app/phant --skill subagent-driven-development-phant-app

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill coordinates multi-task work by dispatching dedicated subagents to implement plan items in sequence, enforcing review gates after each task to ensure quality before progression.

Core Features & Use Cases

  • One-task-at-a-time execution: Dispatches a single implementer subagent per task to avoid context-switching and errors.
  • Two-stage reviews: Automatically assigns a spec-compliance reviewer and a code-quality reviewer; issues must be resolved before moving on.
  • No-commit policy by default: Changes are staged for user approval and are not committed unless explicitly requested.
  • Use Case: In a software project, break down a plan into tasks, have subagents implement each, and pass the work through spec and quality gates to guarantee alignment and quality.

Quick Start

Start by presenting an approved plan, then trigger the subagent-driven execution to begin tasks with mandatory review gates.

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 sequential task execution with subagents in a development workflow?

Subagent-driven development automates sequential task execution by dispatching a single implementer subagent per task to avoid context-switching. Each task is processed one at a time to maintain focus and reduce errors.

What is the best way to enforce code quality and spec compliance during automated task dispatch?

Automated task dispatch enforces quality through two-stage reviews. After a subagent implements a task, a spec-compliance reviewer and a code-quality reviewer check the work, and any issues must be resolved before moving on.

How do I start subagent-driven plan execution for a software project?

To start subagent-driven plan execution, provide an approved plan to trigger the process. The workflow then begins dispatching subagents to implement each task with mandatory review gates.

Does the subagent-driven workflow commit changes automatically after each task?

The subagent-driven workflow uses a no-commit policy by default. Changes are staged for user approval and are not committed unless explicitly requested, ensuring you retain control over the final codebase.

When should I use a plan-driven subagent workflow for task implementation?

Use a plan-driven subagent workflow for complex development tasks that require sequential implementation, plan validation, and quality assurance. It is designed for projects where tasks need dedicated subagent work and strict review gates.

What happens if a subagent's implementation fails the mandatory review gates?

If a subagent's implementation fails the review gates, the issues must be resolved before the workflow can progress to the next task. This ensures quality and spec compliance are maintained throughout the sequence.