subagent-driven-development

Dispatch subagents per task and apply spec compliance and code quality reviews.

3|Updated May 31, 2024
One-click install
npx skills add https://github.com/hugo-berendi/yomi --skill subagent-driven-development-hugo-berendi
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/hugo-berendi/yomi/tree/main/.opencode/skills/subagent-driven-development
Command: npx skills add https://github.com/hugo-berendi/yomi --skill subagent-driven-development-hugo-berendi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill automates plan execution by dispatching a new subagent for each task and enforcing a two-stage review workflow after each task.

Core Features & Use Cases

  • Fresh subagent per task to ensure isolated context and faster iterative feedback.
  • Two-stage review: spec compliance first, then code quality review to guarantee correct implementation.
  • End-to-end orchestration: read the plan, create a TodoWrite with all tasks, dispatch implementers and reviewers, and iterate until all tasks pass.

Quick Start

Use this skill to execute a plan by spawning per-task subagents and applying spec and quality reviews until tasks are completed.

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 plan execution with isolated subagents for each task?

Automating plan execution involves dispatching a new subagent per task to ensure isolated context, creating a TodoWrite with all tasks, and iterating through implementation until completion.

What is a two-stage code review workflow for task dispatch?

A two-stage review workflow checks spec compliance first, then evaluates code quality, ensuring each task passes both review stages before the orchestration proceeds to the next sequential task.

How do I enforce spec compliance before proceeding to the next task in a workflow?

Enforcing spec compliance requires applying a review stage after each task implementation, verifying the output against the plan before allowing the workflow to advance to subsequent tasks.

Do I need a plan file to start automated task orchestration?

Yes, automated task orchestration requires a plan file as input to read the tasks, generate a TodoWrite list, and coordinate the subagents and review stages needed for execution.

Why use a fresh subagent for each task instead of a single agent?

Using a fresh subagent for each task ensures isolated context, preventing interference between sequential tasks and providing faster iterative feedback during the implementation and review process.

What are the limitations of sequential task orchestration with subagents?

Sequential task orchestration processes tasks in isolation one by one, meaning tasks cannot run in parallel, and each must pass both spec compliance and code quality reviews before proceeding.