subagent-driven-development

Dispatch fresh subagents per task with two-stage spec and code quality reviews.

2|Updated Mar 26, 2026
One-click install
npx skills add https://github.com/CUexter/hermes-agent --skill subagent-driven-development-cuexter
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/CUexter/hermes-agent/tree/main/skills/software-development/subagent-driven-development
Command: npx skills add https://github.com/CUexter/hermes-agent --skill subagent-driven-development-cuexter

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates implementation workflows by dispatching independent subagents per task and enforcing a two-stage review process to ensure spec compliance and code quality.

Core Features & Use Cases

  • Fresh per-task subagents maintain clean context and reduce cross-task leakage.
  • Two-stage reviews (spec compliance, then code quality) catch issues early.
  • Plan-driven execution with integrated guidance for TDD and final integration checks.

Quick Start

Load an implementation plan and dispatch the first implementer subagent for the initial 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 automate implementation workflows with strict spec compliance?

Automating implementation workflows with spec compliance requires dispatching fresh subagents per task and enforcing a two-stage review loop to verify requirements before checking code quality.

What is the best way to prevent context leakage between independent coding tasks?

Preventing context leakage between independent coding tasks involves dispatching fresh subagents per task, which maintains clean context separation and reduces cross-task interference during plan-driven execution.

How does a two-stage review process work for TDD and integration testing?

A two-stage review process for TDD works by first checking strict spec compliance, then evaluating code quality, creating a repeatable review loop aligned with integration testing outcomes.

Can I use plan-driven execution for independent tasks that need separate context?

Plan-driven execution supports independent tasks requiring separate context by loading an implementation plan and dispatching dedicated subagents to maintain clean per-task isolation.

When should I not use a subagent delegation workflow for software engineering?

You should not use a subagent delegation workflow for tasks that lack independence or require continuous shared context, because forcing per-task separation disrupts tightly coupled implementation efforts.