subagent-driven-development

Execute implementation plans by dispatching fresh subagents with two-stage reviews.

Updated Dec 22, 2025
One-click install
npx skills add https://github.com/acho1833/spreadline2 --skill subagent-driven-development-acho1833
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/acho1833/spreadline2/tree/main/.claude/commands/subagent-driven-development
Command: npx skills add https://github.com/acho1833/spreadline2 --skill subagent-driven-development-acho1833

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you execute an implementation plan without losing quality by splitting work into independent tasks, assigning each task to a fresh subagent, and reviewing the result before moving on.

Core Features & Use Cases

  • Independent task execution: Break a plan into discrete pieces and handle them one at a time.
  • Two-stage review gate: Validate each task first for spec compliance, then for code quality.
  • Controlled reporting: Use structured prompt templates for implementers and reviewers so every subagent gets the right context.
  • Use case: When you already have a development plan and want to complete it in the same session with strong guardrails and minimal context drift.

Quick Start

Use the subagent-driven-development skill to execute my current implementation plan task by task with a fresh subagent and two review passes per task.

Frequently Asked Questions about subagent-driven-development

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I execute an implementation plan without context drift in a single session?

Prevent context drift during implementation plan execution by dispatching a fresh subagent for each independent task, ensuring sequential validation without context pollution in the same session.

What is the two-stage code review process for task execution?

The two-stage code review process validates each task first for spec compliance, then for code quality, ensuring structured implementer and reviewer prompts enforce sequential task-by-task completion tracking.

How do I automate spec compliance checking during development workflows?

Automate spec compliance checking by applying structured reviewer prompts within a subagent-driven development workflow, validating each task result against the implementation plan before moving on.

Can I use subagent-driven development for tasks that cannot be implemented sequentially?

You should not use subagent-driven development for tasks that cannot be separated, implemented, and validated sequentially, as the workflow requires independent task execution to avoid context pollution.

Do I need a structured implementation plan before starting task execution with subagents?

You need an existing development plan structured into independent tasks before starting, as the workflow requires dispatching a fresh subagent for each discrete piece and reviewing results sequentially.