subagent-driven-development

Dispatch subagents to execute development tasks with automated reviews.

Updated Apr 1, 2026
One-click install
npx skills add https://github.com/evanfang0054/agent-harness --skill subagent-driven-development-evanfang0054
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/evanfang0054/agent-harness/tree/main/skills/subagent-driven-development
Command: npx skills add https://github.com/evanfang0054/agent-harness --skill subagent-driven-development-evanfang0054

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill automates the execution of development plans with independent tasks, streamlining the process with subagent-driven reviews.

Core Features & Use Cases

  • Subagent Execution: Dispatch fresh subagents for each task to maintain task isolation and avoid context pollution.
  • Task Reviews: Automates code reviews for each task to ensure spec compliance and quality.
  • Final Review: Conducts a broad review of the entire branch after all tasks are completed.
  • Use Case: When you have a set of tasks to complete in a development branch, use this Skill to ensure each task is completed correctly and efficiently.

Quick Start

Use the subagent-driven-development skill to execute a plan by providing the task description or Plan path.

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 execution for a development plan without context pollution?

Automate development plan execution by dispatching fresh subagents for each independent task, ensuring task isolation. This subagent-driven approach prevents context pollution while maintaining spec compliance through automated reviews.

How does automated code review work for subagent-driven development?

Automated code review occurs after each subagent executes a task, checking for spec compliance and quality. A final broad review is conducted across the entire development branch once all tasks are completed.

What is the best way to execute independent development tasks in a branch?

The best way to execute independent tasks is using a subagent-driven workflow that isolates each task. This method extracts task briefs from your plan and dispatches subagents sequentially to implement and review code.

How do I set up workflows for subagent-driven development automation?

Setting up subagent-driven development requires configuring scripts for task brief extraction, implementation, and review. You must establish these subagent workflows before providing the task description or plan path.

Can I review an entire development branch after completing all tasks?

Yes, a final broad review of the entire development branch is conducted automatically after all subagent tasks are completed. This ensures overall quality and spec compliance across the full set of implemented tasks.

Do I need to prepare a task description or plan path to start subagent execution?

Yes, you need to provide a task description or plan path to start subagent execution. This input allows the subagent-driven workflow to extract task briefs, dispatch subagents, and automate the required code reviews.