skills.rest
DFly7
DFly7COMMUNITY
Explore all skills by this author →
GitHub

subagent-driven-development

Dispatch isolated subagents per task with two-stage spec and code-quality reviews.

Updated Mar 31, 2026
Search Tags:#automation#tdd#code-review#task-orchestration#spec-compliance#implementer#subagent-driven
One-click install
npx skills add https://github.com/DFly7/iOS-FastAPI-Supabase-AI --skill subagent-driven-development-dfly7
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/DFly7/iOS-FastAPI-Supabase-AI/tree/main/.agents/skills/subagent-driven-development
Command: npx skills add https://github.com/DFly7/iOS-FastAPI-Supabase-AI --skill subagent-driven-development-dfly7

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Coordinating implementation work across multiple independent tasks often leads to context pollution, missed requirements, slow review cycles, and inconsistent quality. This Skill enforces isolated execution for each task by dispatching fresh subagents and applying structured review gates to catch spec and quality issues early, reducing rework and speeding iteration.

Core Features & Use Cases

  • Fresh subagent per task: Launch an implementer subagent with only the exact task context to avoid accidental context leakage.
  • Two-stage review pipeline: Enforce spec compliance review first, then a code-quality review before marking a task done.
  • Model selection and status handling: Choose model capability based on task complexity and handle implementer statuses (DONE, DONE_WITH_CONCERNS, NEEDS_CONTEXT, BLOCKED) to escalate or retry properly.
  • Use Case: Run this when executing a written development plan containing many independent tasks so each task is implemented, tested, and reviewed with minimal human coordination.

Quick Start

Use subagent-driven-development to execute a plan by dispatching an implementer subagent for each task, answering its questions, and running spec and code-quality reviewer subagents after each 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 prevent context pollution when executing multiple development tasks in a single session?▼

To prevent context pollution when executing development tasks, dispatch isolated subagents for each task so they only receive exact task context. This prevents accidental context leakage and ensures each implementation remains focused without interference from other tasks.

What is subagent-driven development and how does it coordinate task execution?▼

Subagent-driven development coordinates task execution by dispatching fresh subagents per task and applying structured two-stage review gates. An implementer subagent executes the work, followed by spec compliance and code-quality reviewer subagents to catch issues early and ensure consistency.

How do I enforce spec compliance and code quality reviews after implementing a task?▼

To enforce spec compliance and code quality reviews, run a two-stage review pipeline after the implementer subagent finishes. Execute the spec reviewer first to verify requirements, then trigger the code-quality reviewer before marking the task complete.

How should I handle implementer subagent statuses like BLOCKED or NEEDS_CONTEXT during task orchestration?▼

Handle implementer subagent statuses like BLOCKED or NEEDS_CONTEXT by escalating or retrying the task appropriately. The workflow processes DONE, DONE_WITH_CONCERNS, NEEDS_CONTEXT, and BLOCKED statuses to coordinate automated task completion and decide whether to provide more context or intervene.

Can I choose different model capabilities for individual tasks within a development plan?▼

Yes, you can choose different model capabilities for individual tasks within a development plan. Model selection guidance allows you to match model capability with task complexity, ensuring appropriate resources are allocated for each dispatched implementer subagent.

When should I use subagent-driven task orchestration instead of manual implementation reviews?▼

Use subagent-driven task orchestration when executing a written development plan containing many independent tasks. It reduces human coordination overhead by automating isolated execution and structured reviews, making it ideal for complex plans requiring strict spec compliance and consistent code quality.

Related Skills

Explore More Software Engineering Skills→
roderikroderik
2

subagent-driven-development

Dispatch fresh subagents per task to execute implementation plans with two-stage reviews.

Community
Advanced
jimmymalhanjimmymalhan
1

rebase-manager

Rebase dependent pull request branches onto main after a base PR merges.

Community
Intermediate
pkuppenspkuppens

v-model-implementation-unit

Align implementation work with unit tests to validate module behavior.

Community
Intermediate
databrickslabsdatabrickslabs
38

subagent-driven-development

Execute multi-task implementation plans with fresh subagents and two-stage reviews.

Official
Advanced