subagent-driven-development

Dispatch fresh subagents to execute tasks from a development plan.

Updated May 23, 2026
One-click install
npx skills add https://github.com/zengbaocheng/hermes-tech-hub --skill subagent-driven-development-zengbaocheng
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/zengbaocheng/hermes-tech-hub/tree/main/software-development/subagent-driven-development
Command: npx skills add https://github.com/zengbaocheng/hermes-tech-hub --skill subagent-driven-development-zengbaocheng

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates plan execution by dispatching fresh subagents for each task, ensuring clean context and higher-quality results across complex workflows.

Core Features & Use Cases

  • Fresh-subagent task execution: each task runs in isolation to avoid context leakage and ensure consistent outcomes.
  • Two-stage review: built-in spec-compliance and code-quality checks before moving to the next task.
  • Flexible plan orchestration: works with plans generated by planning skills and supports automated iteration across tasks in software development pipelines.

Quick Start

Dispatch the first implementer step by providing a plan to delegate_task and initiate the review cycle

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 multi-step plan execution with fresh subagents?

Automate multi-step plan execution by dispatching a fresh subagent for each task to ensure clean context. You provide a plan to a delegate_task interface, which initiates isolated task execution and automated review cycles.

What is a two-stage review in subagent delegation?

A two-stage review is an automated validation process checking spec-compliance and code-quality before completing a delegated task. It ensures each subagent's output meets project standards before moving to the next step.

How do I prevent context leakage when running complex development workflows?

Prevent context leakage in complex workflows by isolating each task in a fresh subagent. Running each step with clean context avoids residual information from previous tasks and produces consistent, high-quality outcomes.

Can I use this subagent workflow with plans generated by other planning skills?

Yes, you can use plans generated by other planning skills. This workflow supports flexible plan orchestration by accepting a plan input and automating task iteration across software development pipelines.

What is the best way to ensure code quality in automated task delegation?

Ensure code quality in automated task delegation by using a workflow with built-in spec and quality checks. Dispatching fresh subagents per task followed by a two-stage review guarantees consistent, high-quality results.