subagent-driven-development

Execute implementation plans by dispatching subagents with two-stage reviews.

11|4|Updated Feb 9, 2026
One-click install
npx skills add https://github.com/wpank/ai --skill subagent-driven-development-wpank
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/wpank/ai/tree/main/skills/tools/subagent-driven-development
Command: npx skills add https://github.com/wpank/ai --skill subagent-driven-development-wpank

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill automates the execution of implementation plans by breaking them down into discrete tasks, each handled by a dedicated subagent with a two-stage review process, ensuring high quality and fast iteration without manual intervention.

Core Features & Use Cases

  • Task Decomposition: Automatically extracts all tasks from a given plan.
  • Subagent Dispatch: Assigns each task to a specialized 'Implementer' subagent.
  • Two-Stage Review: Ensures spec compliance first, then code quality, before marking a task complete.
  • Use Case: You have a detailed plan to refactor a module. This Skill can autonomously execute each refactoring step, ensuring each change meets the specification and adheres to coding standards before moving to the next.

Quick Start

Execute the implementation plan located at '/path/to/your/plan.md' using the subagent-driven development skill.

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 implementation plan execution for discrete development tasks?

Automating implementation plan execution is achieved by extracting discrete tasks and dispatching a fresh subagent per task. This approach ensures fast iteration and high-quality results without manual intervention.

What is the two-stage review process for autonomous code generation?

The two-stage review process for code generation enforces spec compliance first, then evaluates code quality. A specialized subagent must pass both stages before the system marks the task complete.

How do I execute a refactoring plan autonomously within a single development session?

You can execute a refactoring plan autonomously by providing the plan file path. The system automatically decomposes the plan into independent tasks and manages subagent execution and review within the session.

Can I use subagents to handle independent tasks requiring strict spec compliance?

Yes, subagents are designed for independent tasks requiring strict spec compliance. The system provides necessary context to each subagent and enforces a strict review order to verify adherence.

When should I not use subagent-driven development for my workflow?

You should not use subagent-driven development for plans lacking discrete, independent tasks. It is specifically designed for high-quality, fast iteration of well-defined tasks within a single session.