skills.rest
DYAI2025
DYAI2025COMMUNITY
Explore all skills by this author →
GitHub

subagent-driven-development

Dispatch fresh subagents for each task with inter-task code reviews.

3|Updated Oct 18, 2025
One-click install
npx skills add https://github.com/DYAI2025/Stoppclock-page --skill subagent-driven-development-dyai2025
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/DYAI2025/Stoppclock-page/tree/main/stoppclock_speckit/.claude/commands/SKILL (22).md
Command: npx skills add https://github.com/DYAI2025/Stoppclock-page --skill subagent-driven-development-dyai2025

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill optimizes the execution of implementation plans by dispatching fresh subagents for each task, followed by a code review. This approach ensures high-quality code, fast iteration, and early detection of issues, preventing context pollution and costly rework.

Core Features & Use Cases

  • Task-Specific Subagents: Dispatch a new subagent for each task, ensuring focused execution and preventing context drift.
  • Inter-Task Code Review: Integrate code reviews between tasks to catch issues early and maintain quality.
  • Use Case: Execute a complex feature implementation plan, leveraging multiple AI agents to work through tasks efficiently while maintaining strict quality gates and continuous oversight.

Quick Start

1. Load Plan: Read plan file, create TodoWrite with all tasks.

2. Execute Task with Subagent: Dispatch fresh subagent for a task.

3. Review Subagent's Work: Dispatch code-reviewer subagent.

4. Apply Review Feedback: Fix issues found by reviewer.

5. Mark Complete, Next Task: Repeat until all tasks are done.

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 with multiple independent tasks efficiently?▼

Subagent-driven development dispatches a fresh subagent for each task, enabling focused execution and fast iteration. This approach isolates tasks to prevent context drift and maintains code quality through mandatory inter-task code reviews before progressing to the next task.

What's the best way to maintain code quality when working through a complex feature plan?▼

Integrate code reviews between tasks by dispatching a dedicated code-reviewer subagent after each task completes. This catches issues early, prevents rework, and ensures each task meets quality standards before the next one begins, keeping the codebase merge-ready.

How do I prevent context pollution when executing multiple implementation tasks?▼

Per-task subagent instantiation isolates each task's context, preventing accumulated state or decisions from one task from affecting another. Each subagent works independently, then passes reviewed, validated code to the next task in the workflow.

Can I run test-driven development as part of task execution?▼

Yes. Subagent-driven development satisfies test-driven development requirements where needed by enforcing task-level implementation with tests. Each subagent handles both implementation and test creation, ensuring quality gates are met per task.

What happens after a subagent completes a task?▼

The code-reviewer subagent inspects the completed work, identifies issues, and provides feedback. You apply the review findings, then mark the task complete and proceed to dispatch a fresh subagent for the next task in the plan.

Do I need to manually load the implementation plan before starting?▼

Yes. Load the plan file and create a todo list with all tasks before execution begins. This initial setup enables the workflow to dispatch subagents sequentially and track progress through to final merge-ready validation.

Related Skills

szaherszaher

speckit-plan

Generate technical implementation plans from feature specifications into plan.md.

Community
Advanced
blakesimsblakesims

review-merge

Validate feature branch merge readiness with integration checks and summaries.

Community
Advanced
brabos-aibrabos-ai
8

add-subagent-driven-development

Dispatch independent subagents for iterative development tasks with enforced code reviews.

Official
Advanced
micaelmaltamicaelmalta
1

workflow

Coordinate multi-phase software development workflows from plan to GitHub PR.

Community
Advanced