skills.rest
bearhedge
bearhedgeCOMMUNITY
Explore all skills by this author →
GitHub

subagent-driven-development

Dispatch independent subagents for isolated task implementation with two-stage review.

Updated Aug 23, 2026
Search Tags:#code quality#TDD#subagents#plan execution#spec compliance#task orchestration#two-stage review
One-click install
npx skills add https://github.com/bearhedge/APE-YOLO --skill subagent-driven-development-bearhedge
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/bearhedge/APE-YOLO/tree/main/.claude/skills/subagent-driven-development
Command: npx skills add https://github.com/bearhedge/APE-YOLO --skill subagent-driven-development-bearhedge

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill enables executing complex implementation plans by dispatching fresh subagents for each task within the same session, ensuring isolated work streams, rapid feedback, and structured coordination.

Core Features & Use Cases

  • Fresh per-task subagents for clean context and zero cross-task contamination
  • Two-stage review: first verify spec compliance, then assess code quality
  • Controlled workflow: per-task execution with clear handoffs and iterative refinement

Quick Start

  • Load your plan and announce the task list.
  • Dispatch an implementer subagent for Task N with full task text and context.
  • After implementation, dispatch a spec compliance reviewer to validate alignment with the plan.
  • If issues are found, have the implementer fix and re-run the reviews.
  • Dispatch the code quality reviewer; repeat fixes if needed until approved.
  • Mark Task N complete and proceed to the next task or finish the plan.

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 multiple independent tasks in a single session without cross-task context contamination?▼

Subagent-driven development executes independent tasks by dispatching fresh subagents for each task within a single session, ensuring isolated work streams and zero cross-task context contamination.

What is a two-stage review workflow for task orchestration?▼

A two-stage review workflow first dispatches a spec compliance reviewer to validate alignment with the plan, then dispatches a code quality reviewer to assess implementation standards before marking the task complete.

How do I orchestrate plan execution and enforce spec compliance for complex implementations?▼

You orchestrate plan execution by loading a predefined plan, dispatching an implementer subagent for each task, and running iterative spec compliance and code quality reviews until the task passes both stages.

Do I need predefined prompts and templates to use subagents for task isolation?▼

Yes, this task isolation approach requires predefined prompts and templates for the implementer, spec reviewer, and code-quality reviewer roles to integrate them into a disciplined orchestration process.

When should I not use fresh subagents for plan execution?▼

You should avoid subagent-driven development for plans lacking multiple independent tasks, as its core mechanism relies on dispatching isolated subagents for largely independent task execution and structured handoffs.

Related Skills

Explore More Software Engineering Skills→
mistakeknotmistakeknot
2

subagent-driven-development

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

Community
Advanced
m16khbm16khb

subagent-driven-development

Dispatch fresh subagents per task with code review and test gates.

Community
Advanced
Coffelix2023Coffelix2023
1

subagent-driven-development

Dispatch AI subagents for each task in an implementation plan.

Community
Advanced
mrtolkienmrtolkien
5

subagent-development

This is a stub. Replace with actual content.

Community
Advanced