subagent-driven-development

Name: subagent-driven-development
Availability: InStock
Author: charly3pins

Dispatch fresh subagents per task with isolated contexts and post-task code reviews.

4|Updated Dec 7, 2020

Search Tags:#code review #development workflow #AI development #iteration #task execution #quality gates #subagent

Ready to run

npx skills add https://github.com/charly3pins/dotfiles --skill subagent-driven-development-charly3pins

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/charly3pins/dotfiles/tree/main/.config/opencode/skills/subagent-driven-development
Command: npx skills add https://github.com/charly3pins/dotfiles --skill subagent-driven-development-charly3pins

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Executing complex implementation plans can lead to context pollution, late-stage bug discovery, and slow iteration if not managed with continuous quality checks and focused work.

Core Features & Use Cases

Fresh Subagent per Task: Dispatches a new, focused subagent for each task, preventing context pollution and ensuring dedicated work.
Inter-Task Code Review: Integrates automated code review after each task, catching issues early and maintaining high quality throughout development.
Fast Iteration: Enables continuous progress within the same session, reducing handoff delays and accelerating the development cycle.
Use Case: When you have an implementation plan with several independent tasks, use this skill to execute them sequentially, with an automated code review after each task to ensure high quality and rapid iteration.

Quick Start

I have an implementation plan. Use subagent-driven development to execute the tasks, with code review after each one.

Frequently Asked Questions about subagent-driven-development

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

FAQPage Schema

How do I prevent context pollution when executing multiple development tasks in sequence?▼

Context pollution occurs when a single agent accumulates state across independent tasks, causing early decisions to constrain later work and bugs to compound. Subagent-driven development dispatches a fresh, isolated subagent for each task, preventing carryover and enabling focused execution with clean reasoning.

What's the best way to catch code quality issues early during iterative development?▼

Integrate automated code review after each task completes rather than waiting for final review. Subagent-driven development embeds per-task code reviews into the workflow, catching issues immediately and maintaining quality throughout iteration instead of discovering problems late.

Can I execute an implementation plan with independent tasks faster using quality gates?▼

Yes. Subagent-driven development executes tasks sequentially with built-in quality gates—code review and test-driven validation after each task—reducing handoff delays and enabling continuous progress within the same session while maintaining high standards.

When should I use isolated task execution instead of a single agent for my development workflow?▼

Use isolated execution when your implementation plan contains several independent tasks that don't require accumulated context from prior work. This approach prevents context pollution, accelerates iteration, and is applicable when tasks can be executed in sequence with post-task reviews.

What requirements do I need to meet to use subagent-driven development effectively?▼

You need per-task subagent creation capability, isolated execution contexts, post-task code review integration, test-driven validation, and orchestration supporting plan loading, task tracking, and final review. The workflow requires independent tasks within the same session.