subagent-driven-development

Dispatch fresh implementer and reviewer subagents per task with TDD checks.

1|Updated Apr 12, 2026
One-click install
npx skills add https://github.com/cryptopafi/nexusos-skills --skill subagent-driven-development-cryptopafi
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/cryptopafi/nexusos-skills/tree/main/software-development/subagent-driven-development
Command: npx skills add https://github.com/cryptopafi/nexusos-skills --skill subagent-driven-development-cryptopafi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill orchestrates per-task implementation and two-stage automated reviews to prevent context pollution, catch specification drift early, and maintain consistent code quality while enabling parallel progress.

Core Features & Use Cases

  • Fresh subagent per task: dispatches isolated implementer subagents so each task runs in a clean context without leftover state.
  • Two-stage review workflow: enforces automated spec compliance checks followed by a dedicated code quality review before marking tasks complete.
  • TDD-aligned implementation and final integration review: requires failing-first tests, iterative fix-and-review loops, and a final integration reviewer to ensure all tasks work together in the full test suite.
  • Use Case: Given a feature plan broken into independent tasks, this Skill automates creating a todo list, dispatching implementers and reviewers for each task, cycling fixes until approvals, and producing a merge-ready implementation.

Quick Start

Dispatch the parsed implementation plan to create a todo list and iterate implementer, spec reviewer, and quality reviewer subagents per task until all tasks pass and integrate.

Frequently Asked Questions about subagent-driven-development

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

FAQPage Schema
How do I automate code review and implementation for independent tasks?

Automating code review and implementation involves dispatching fresh implementer and reviewer subagents per task to execute plans in isolated contexts. This enforces automated spec compliance and code quality reviews before marking tasks complete.

What is the best way to prevent context pollution during parallel software development?

Preventing context pollution during parallel software development is achieved by spawning fresh, isolated subagents for each task. This ensures every task runs in a clean context without leftover state affecting the implementation.

How do I enforce TDD and spec compliance checks in automated task delegation?

Enforcing TDD and spec compliance in automated task delegation requires failing-first tests, iterative fix-and-review loops, and a dedicated spec reviewer. A final integration reviewer then runs the full project test suite to ensure all tasks work together.

Can I use subagent-driven development for features without independent tasks?

Subagent-driven development is designed for software development projects with independent tasks. If your feature plan cannot be broken down into independent units, the parallel subagent orchestration and fresh context isolation will not function effectively.

Why does automated task delegation require a two-stage review workflow?

A two-stage review workflow is required to catch specification drift early and maintain consistent code quality. It enforces automated spec compliance checks first, followed by a dedicated code quality review before marking tasks complete.

Do I need to provide full task text when dispatching implementer subagents?

Providing full task text in each subagent context is required when dispatching delegate_task workers. This ensures the isolated implementer subagents have all necessary specifications to execute the task without relying on external state.