subagent-driven-development

Execute implementation plans by delegating tasks to subagents with spec and quality reviews.

4|Updated May 18, 2026
One-click install
npx skills add https://github.com/ZardLi1115/zedclaw --skill subagent-driven-development-zardli1115
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/ZardLi1115/zedclaw/tree/main/skills/software-development/subagent-driven-development
Command: npx skills add https://github.com/ZardLi1115/zedclaw --skill subagent-driven-development-zardli1115

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps you turn an implementation plan into correct, high-quality code by running isolated task executors and enforcing two-stage verification before anything proceeds.

Core Features & Use Cases

  • Per-task delegate_task execution: Creates a fresh implementer subagent for each task to reduce context pollution and keep work focused.
  • Two-stage review pipeline: Runs a spec compliance reviewer first, then a code quality reviewer, looping back to fixes until approved.
  • TDD-aligned implementation: Instructs implementer subagents to write failing tests first, implement minimal changes, and verify no regressions.
  • Integration review after completion: Performs a final end-to-end review of consistency and test health across all tasks.

Quick Start

Run the subagent-driven-development skill to execute a tasks plan by dispatching an implementer, validating spec compliance, and then validating code quality for each 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 subagents for multi-task software implementation?

Subagent orchestration delegates each implementation task to a fresh subagent, reducing context pollution and enforcing spec-first then quality-first review loops until all tasks pass.

Can I enforce test-driven development when delegating tasks to subagents?

Yes, subagent delegation supports TDD alignment by instructing implementer subagents to write failing tests first, implement minimal changes, and verify no regressions before proceeding.

How does spec compliance review work for isolated task executors?

Spec compliance review runs as the first stage of a two-stage pipeline, validating the implementer subagent's output against requirements before passing it to a code quality reviewer for final approval.

What is the best way to verify code consistency after completing independent multi-task changes?

Final integration verification performs an end-to-end review of consistency and test health across all completed tasks, ensuring deterministic progress and reliable verification for multi-task changes.

When should I use a two-stage review pipeline with subagent delegation?

Use subagent delegation with a two-stage review pipeline for independent, multi-task software changes where you need deterministic progress, reduced context pollution, and reliable verification.

Do I need to provide complete task context up front for subagent workflow orchestration?

Yes, subagent workflow orchestration requires per-task delegate_task execution with complete task context provided up front to ensure bounded revision loops until PASS or APPROVED status.