subagent-driven-development

Dispatch per-task subagents to implement plan items with isolated context.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Fresh per-task subagents to implement plan items with isolated context, ensuring clean separation and traceable progress.

Core Features & Use Cases

  • Fresh per-task subagents implement plan items independently, reducing context leakage and cross-task interference.
  • Two-stage review per task: spec-compliance reviewer and code-quality reviewer, plus a final integration check.
  • Plan parsing and per-task orchestration with structured feedback loops to enable fast, reliable iteration.

Quick Start

Dispatch a project plan into per-task subagents and initiate the two-stage review workflow to complete tasks.

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 delegation to subagents for isolated task implementation?

Automating code delegation involves dispatching per-task subagents to implement plan items with isolated context, ensuring clean separation and traceable progress without cross-task interference.

What is the two-stage review process for subagent-driven development?

The two-stage review process applies a spec-compliance check followed by a code-quality check per task, plus a final integration check to ensure task completion matches the original plan.

How do I parse a project plan into subagent tasks with structured feedback loops?

Parsing a project plan into subagent tasks uses plan parsing and per-task orchestration with structured feedback loops, enabling fast and reliable iteration through isolated execution.

Why does context leakage happen when automating multi-task workflows, and how can I prevent it?

Context leakage happens when tasks share execution environments; preventing it requires dispatching fresh per-task subagents that implement plan items independently with isolated context.

Can I use subagent orchestration for multi-step plan execution without external dependencies?

Subagent orchestration runs without external dependencies, using built-in plan parsing and per-task delegation to execute multi-step plans with isolated context and two-stage reviews.

What is the best way to ensure spec compliance when delegating coding tasks to subagents?

Ensuring spec compliance involves applying a two-stage review per task, where a spec-compliance reviewer checks alignment before a code-quality reviewer assesses the implementation.