subagent-driven-development

Dispatch fresh subagents per task with two-stage reviews.

2|1|Updated May 1, 2026
One-click install
npx skills add https://github.com/LeonardoTan19/metaforge --skill subagent-driven-development-leonardotan19
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/LeonardoTan19/metaforge/tree/main/skills/subagent-driven-development
Command: npx skills add https://github.com/LeonardoTan19/metaforge --skill subagent-driven-development-leonardotan19

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Subagent-driven development helps orchestrate plan execution by delegating each task to a fresh subagent with isolated context, ensuring clean, deterministic progress.

Core Features & Use Cases

  • Dispatch implementer subagents per task with two-stage reviews (spec then quality)
  • Maintain session isolation to avoid context leakage while enabling rapid iteration
  • Finish with a consolidated development branch after all tasks are complete

Quick Start

Dispatch the first implementer subagent for the initial task and verify results.

Frequently Asked Questions about subagent-driven-development

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

FAQPage Schema
How do I orchestrate task plan execution with autonomous coding agents?

Subagent-driven development orchestrates plan execution by dispatching a fresh subagent per task and applying two-stage reviews after each task. It ensures clean, deterministic progress through automated implementer and reviewer subagents.

What is subagent-driven development and how does it handle context isolation?

Subagent-driven development delegates each task to a fresh subagent with isolated context. This mechanism maintains session isolation to avoid context leakage while enabling rapid iteration through automated implementer and reviewer subagents.

How do I automate two-stage spec and code quality reviews in my development workflow?

You can automate two-stage reviews by applying spec and code quality checks after each task. The workflow relies on spec-reviewer and code-quality-reviewer prompts to evaluate the implementer subagent's output before proceeding.

Can I execute independent tasks in parallel using isolated subagents?

Yes, subagent-driven development is applied to sessions with independent tasks. It enables rapid iteration by dispatching a fresh implementer subagent per task while maintaining isolated context for each execution.

How do I consolidate development branches after completing all subagent tasks?

You can consolidate changes using a finishing branch step after all tasks are complete. This step consolidates the work done by the implementer and reviewer subagents into a single development branch.