subagent-driven-development

Dispatch fresh subagents for each plan task with two-stage reviews.

Updated Mar 30, 2026
One-click install
npx skills add https://github.com/steltz/skill-factory --skill subagent-driven-development-steltz
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/steltz/skill-factory/tree/main/references/superpowers/skills/subagent-driven-development
Command: npx skills add https://github.com/steltz/skill-factory --skill subagent-driven-development-steltz

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Subagent-Driven Development enables executing implementation plans by dispatching dedicated subagents for each task in a single session, with isolated context and structured review.

Core Features & Use Cases

  • Two-stage review workflow per task (spec compliance first, then code quality) to ensure correctness and quality.
  • Per-task delegation using independent subagents to keep tasks loosely coupled and traceable.
  • Clear escalation and re-dispatch when blockers arise to maintain momentum.

Quick Start

Register the skill and start dispatching implementer prompts per plan task.

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 independent tasks in a single session using subagents?

You can orchestrate task execution by dispatching a fresh subagent for each implementation task, keeping the tasks loosely coupled and traceable. The controller manages task creation, reviews, and final integration.

What is the two-stage review workflow for code task execution?

The two-stage review workflow checks spec compliance first, then evaluates code quality between tasks. This structured approach ensures both correctness and quality before proceeding to the next implementation step.

How do I handle blockers during subagent-driven task orchestration?

When blockers arise during task execution, the workflow uses clear escalation and re-dispatch of subagents to maintain momentum. This ensures the implementation plan continues progressing without getting stalled.

What prompts do I need to set up subagent-driven development?

You need dedicated prompts for the implementer, spec reviewer, and code-quality reviewer. These prompts define the isolated context and structured review process for each dispatched subagent.

When should I use subagent-driven development for plan execution?

Use this approach for implementation plans that contain mostly independent tasks. It is designed for single-session execution where keeping tasks loosely coupled and traceable through isolated subagent context is beneficial.

Does subagent-driven development require any external dependencies?

No external dependencies are required. The skill operates independently by using a controller to orchestrate task creation, dispatch subagents, manage the two-stage review workflow, and perform final integration.