subagent-driven-development

Delegate plan tasks to fresh subagents with two-stage spec and code reviews.

Updated May 21, 2026
One-click install
npx skills add https://github.com/samlaggz/hermes-agent-sam --skill subagent-driven-development-samlaggz
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/samlaggz/hermes-agent-sam/tree/main/backup/skills/software-development/subagent-driven-development
Command: npx skills add https://github.com/samlaggz/hermes-agent-sam --skill subagent-driven-development-samlaggz

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill enables orchestration of plan execution by delegating implementation tasks to fresh subagents, followed by two-stage reviews to ensure specification alignment and code quality.

Core Features & Use Cases

  • Fresh per-task subagents to avoid context bleed and ensure focused reasoning.
  • Two-stage review workflow: spec compliance reviewer and code quality reviewer before integration.
  • End-to-end task execution: parse a plan, dispatch tasks, validate outcomes, and commit upon approval.
  • Guardrails include explicit context propagation, stepwise task decomposition, and revision-loop safety checks.

Quick Start

Initiate Task 1 by dispatching an implementer subagent with full context of the plan and requirements.

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 task delegation to subagents for plan execution?

Automate task delegation by dispatching fresh subagents per task, ensuring focused reasoning without context bleed. A structured two-stage review workflow validates specification compliance and code quality before final integration and commit.

Why use fresh subagents for each task instead of a single context window?

Fresh subagents prevent context bleed, ensuring focused reasoning for each task. This approach isolates implementation logic, which is then validated through sequential spec compliance and code quality reviews before integration.

How does the two-stage code review workflow validate spec compliance?

The two-stage review workflow validates tasks sequentially: first checking specification alignment, then evaluating code quality. This structured review process enforces iterative refinement and checkpoint-based retries before the final integration commit.

Can I use subagent orchestration for plans with dependent tasks and unclear specifications?

Subagent orchestration is applicable to plans with independent tasks and clear specifications. It enforces a plan reading step and stepwise task decomposition, making it unsuitable for unclear specifications or tightly dependent task structures.

What's the best way to orchestrate implementation tasks with iterative refinement?

Orchestrate iterative refinement by delegating tasks to fresh subagents, followed by sequential spec compliance and code quality reviews. The workflow includes a final integration checkpoint with revision-loop safety checks and checkpoint-based retries.

Does subagent delegation support automated per-task execution and integration checkpoints?

Subagent delegation supports automated per-task execution, sequential reviews, and a final integration checkpoint. It enforces context propagation and checkpoint-based retries to ensure outcomes meet specifications before committing.