🎯

subagent-driven-development

🎯Skill

from obra/superpowers

What it does

Automates complex software development workflows by dynamically coordinating specialized AI subagents to break down and solve intricate programming tasks

Overview

A Claude Code skill for subagent-driven development from the Superpowers collection by obra. It automates complex software development workflows by dispatching fresh AI subagents for each task, with a two-stage review process (spec compliance followed by code quality) to ensure high-quality output.

Key Features

Fresh subagent dispatch per task to prevent context pollution
Two-stage review: spec compliance review then code quality review
Automated implementation plan execution with TodoWrite tracking
Structured implementer, spec reviewer, and code quality reviewer roles
Fast iteration without human-in-loop between tasks

Who is this for?

Developers working on complex implementation plans with multiple independent tasks who want automated, high-quality execution. Ideal for those who have a clear plan and want Claude Code to systematically implement each task with built-in quality assurance through the two-stage review process.

📦

Same repository

obra/superpowers(16 items)

📚 Related Resources

📝Superpowers: How I'm using coding agents in October 2025

blog.fsck.com

📰Simon Willison's Notes on Superpowers

simonwillison.net

subagent-driven-development

Installation

Vibe Index InstallInstalls to .claude/skills/ - auto-recognized by Claude Code

npx vibeindex add obra/superpowers --skill subagent-driven-development

skills.sh Install⚠ Installs to .agents/skills/ - may not be auto-recognized by Claude Code

npx skills add obra/superpowers --skill subagent-driven-development

Manual InstallCopy SKILL.md content and save to the path below

~/.claude/skills/subagent-driven-development/SKILL.md

SKILL.md

18,530Installs

43,383

AddedJan 31, 2026

View on GitHub Back to Skills

More from this repository10

🏪

obra-superpowers🏪Marketplace

Complete software development workflow framework for coding agents. Enforces structured processes — brainstorming, spec writing, TDD, subagent-driven development, and code review — turning AI agents into disciplined engineering teammates.

🎯

brainstorming🎯Skill

Generates structured, creative problem-solving frameworks with step-by-step ideation techniques, prioritization, and actionable innovation strategies.

🎯

systematic-debugging🎯Skill

Methodically diagnose and resolve complex software bugs using structured analysis, trace patterns, and strategic troubleshooting techniques.

🎯

test-driven-development🎯Skill

Guides developers in writing automated tests before code, ensuring robust, modular software design and catching bugs early in the development process.

🎯

writing-plans🎯Skill

Generates structured, actionable writing plans with clear goals, research strategies, outline templates, and milestone tracking for various document types.

🎯

requesting-code-review🎯Skill

Streamlines code review requests by generating structured feedback templates, tracking review status, and facilitating collaborative code improvement workflows.

🎯

using-superpowers🎯Skill

Streamlines integration and usage of superpowers framework, providing intuitive methods for defining, configuring, and applying advanced computational capabilities.

🎯

executing-plans🎯Skill

Transforms complex multi-step workflows into executable sequences with robust error handling, retry mechanisms, and intelligent task orchestration.

🎯

receiving-code-review🎯Skill

Helps developers professionally receive, process, and learn from code review feedback to improve code quality and collaboration skills.

🎯

verification-before-completion🎯Skill

Validates and cross-checks critical task parameters, inputs, and preconditions before finalizing complex workflows to prevent errors and ensure reliable execution.