skills.rest
jnMetaCode
jnMetaCodeCOMMUNITY
Explore all skills by this author →
𝕏@jnMetaCodeGitHub

subagent-driven-development

Delegate independent tasks to fresh subagents with two-stage reviews.

7.6k|721|Updated Mar 20, 2026
Search Tags:#automation#planning#delegation#subagent#two-stage-review#task-distribution#ai-workflow
One-click install
npx skills add https://github.com/jnMetaCode/superpowers-zh --skill subagent-driven-development-jnmetacode
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/jnMetaCode/superpowers-zh/tree/main/skills/subagent-driven-development
Command: npx skills add https://github.com/jnMetaCode/superpowers-zh --skill subagent-driven-development-jnmetacode

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill enables fast, high-quality task execution by assigning each task to a fresh, isolated subagent, preventing context bleed and enabling parallel work with clear review gates.

Core Features & Use Cases

  • Isolated subagents for each task to avoid cross-task context contamination.
  • Two-stage review: first verify the plan/specs, then verify code quality.
  • Structured workflow that supports parallel execution within a single session and rapid iteration.

Quick Start

Start by delegating the first task to a new subagent and initiate the two-stage review process.

Frequently Asked Questions about subagent-driven-development

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I prevent context bleed when distributing independent coding tasks to multiple AI agents?▼

To prevent context bleed, delegate each independent coding task to a fresh, isolated subagent within your current session. This ensures parallel execution without cross-task context contamination, maintaining clear boundaries and focused task execution.

What is a two-stage review process for AI workflow automation and when should I apply it?▼

A two-stage review process verifies plan specifications first, then evaluates code quality. Apply this AI workflow automation technique to plans containing multiple independent tasks to ensure safety and quality before finalizing execution.

How do I delegate tasks to subagents for fast results without compromising code safety?▼

Delegate tasks to isolated subagents for rapid iteration, then enforce a structured two-stage review process. This workflow verifies the initial plan specs and subsequent code quality, ensuring safety while supporting fast, parallel execution.

Do I need a specific file format to set up task delegation for subagents?▼

Yes, task delegation requires a SKILL.md frontmatter containing the name and description. You can also include optional scripts, references, and assets alongside the provided prompt templates to govern subagent execution.

Can I run multiple independent subagents in parallel within a single session?▼

Yes, you can run multiple independent subagents in parallel within a single session. This structured workflow enables rapid iteration by assigning each task to a fresh subagent, preventing context bleed while accelerating execution.

Related Skills

Explore More Software Engineering Skills→
lagz0nelagz0ne
2

subagent-driven-development

Dispatch fresh subagents per task with two-stage spec and code reviews.

Community
Advanced
Strategic-AutomationStrategic-Automation
6

subagent-driven-development

Dispatch independent subagents per task with two-stage reviews before integration.

Official
Advanced
cruldracruldra

subagent-driven-development

Delegate each task to a fresh subagent and apply two-stage reviews.

Community
Advanced
charly3pinscharly3pins
4

subagent-driven-development

Dispatch fresh subagents per task with isolated contexts and post-task code reviews.

Community
Basic