subagent-driven-development

Dispatch a fresh subagent per task and run two-stage reviews.

47|11|Updated Feb 25, 2026
One-click install
npx skills add https://github.com/tylerwind/LabVIEW-WPF-UI-Generator --skill subagent-driven-development-tylerwind
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/tylerwind/LabVIEW-WPF-UI-Generator/tree/main/.agent/skills/subagent-driven-development
Command: npx skills add https://github.com/tylerwind/LabVIEW-WPF-UI-Generator --skill subagent-driven-development-tylerwind

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This capability eliminates bottlenecks in complex plan execution by dividing work among fresh subagents per task, ensuring clean context per task and rapid iteration.

Core Features & Use Cases

  • Per-Task Subagents: spawn a dedicated subagent for each task to maintain isolation and reduce cross-task context pollution.
  • Two-Stage Review: require a spec compliance review before any code-quality review to prevent scope creep and regression.
  • Automatic Progress Tracking: mark tasks complete after successful reviews and move to the next task seamlessly.
  • Use Case: In a multi-step implementation plan, assign independent tasks to subagents and validate through two-stage reviews before merging.

Quick Start

Dispatch a new subagent for each plan task, perform spec-compliance review, then code-quality review, and mark tasks complete before proceeding.

Frequently Asked Questions about subagent-driven-development

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

FAQPage Schema
How do I isolate context when executing multiple independent coding tasks in a single planning session?

To isolate context for independent tasks, you can dispatch a fresh subagent per task within your current planning session. This approach maintains clean context per task, preventing cross-task context pollution and enabling rapid iteration.

What is the best way to automate spec compliance and code quality reviews for each task in a development plan?

Automating spec compliance and code quality reviews is best handled through a two-stage review process per task. This requires a spec-compliance review to prevent scope creep, followed strictly by a code-quality review before marking the task complete.

How do I automatically track completion status for multi-step implementation plans?

You can automatically track task completion by dispatching subagents per plan task and enforcing two-stage reviews. The system marks tasks complete after successful reviews and seamlessly moves to the next task in the workflow.

Can I use subagent task dispatch for dependent tasks within an implementation plan?

Subagent task dispatch is applicable for largely independent tasks in a single planning session. It enforces per-task isolation, so it is not suited for tasks with heavy dependencies that require shared context across steps.

Why does my complex plan execution bottleneck when using a single agent context?

Complex plan execution bottlenecks in a single context due to accumulated context pollution. Dividing work among fresh subagents per task eliminates this bottleneck by ensuring clean context and rapid iteration for each step.