skills.rest
tyk-lab
tyk-labCOMMUNITY
Explore all skills by this author →
GitHub

subagent-driven-development

Automate multi-task plan execution with fresh subagents and two-stage reviews.

Updated Feb 5, 2026
Search Tags:#workflow#development#subagent#ai-agent#task-management#code-review#prompt-engineering
One-click install
npx skills add https://github.com/tyk-lab/my-ai-skill --skill subagent-driven-development-tyk-lab
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/tyk-lab/my-ai-skill/tree/main/skills/subagent-driven-development
Command: npx skills add https://github.com/tyk-lab/my-ai-skill --skill subagent-driven-development-tyk-lab

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates plan execution by distributing tasks to fresh subagents, with a two-stage review after each task to enforce spec compliance and high-quality outcomes.

Core Features & Use Cases

  • Fresh subagent per task to avoid context pollution and enable rapid iteration
  • Two-stage reviews: spec compliance followed by code quality
  • Works within a single session for interdependent planning and execution

Quick Start

Initiate a plan by dispatching a fresh subagent per task and run two-stage reviews 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 automate multi-task plan execution with AI agents while avoiding context pollution?▼

Automate multi-task plan execution by dispatching a fresh subagent for each task. This approach prevents context pollution, enabling rapid iteration within a single session while maintaining independent task isolation.

What is the two-stage review process for subagent-driven development?▼

The two-stage review process enforces spec compliance and code quality after each task. A fresh subagent implements the task, followed by dual review templates evaluating specification adherence and overall code quality.

How do I manage independent coding tasks for rapid iteration within a single session?▼

Manage independent coding tasks by distributing them to fresh subagents within a single session. This workflow requires a plan source and per-task implementer prompts to execute rapid iterations effectively.

Does subagent-driven development work for tightly interdependent task plans?▼

Subagent-driven development is designed for plans where tasks are largely independent. It applies to multi-task plans requiring rapid iteration, using fresh subagents to avoid context pollution during single session execution.

What do I need to set up before dispatching subagents for task execution?▼

You need to provide a plan source, per-task implementer prompts, and dual review templates. These inputs enforce spec compliance and code quality during the two-stage review after each task execution.

Related Skills

Explore More Software Engineering Skills→
alffeialffei
69

subagent-driven-development

Dispatch fresh subagents per task with isolated context and two-stage reviews.

Community
Advanced
galangryandanagalangryandana
20

subagent-driven-development

Dispatch fresh subagents to execute independent implementation tasks with two-stage reviews.

Community
Advanced
VetCodersVetCoders
1

vetcoders-implement

Delegate AI subtasks to native Claude agents using the built-in Task tool.

Official
Intermediate
pproencapproenca

subagent-driven-development

Execute implementation plans by dispatching a fresh subagent per task.

Community
Basic