subagent-driven-development

Orchestrate plan execution by dispatching fresh subagents per task with two-stage automated reviews.

8|Updated Sep 7, 2021
One-click install
npx skills add https://github.com/frap129/dotfiles --skill subagent-driven-development-frap129
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/frap129/dotfiles/tree/main/.config/opencode/skills/subagent-driven-development
Command: npx skills add https://github.com/frap129/dotfiles --skill subagent-driven-development-frap129

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Use Subagent-Driven Development to orchestrate plan execution by dispatching fresh subagents per task within the same session. This approach enables isolated task handling, reduces context pollution, and accelerates iteration while maintaining tight control over quality through structured reviews.

Core Features & Use Cases

  • Fresh subagent per task plus a two-stage review (spec compliance first, then code quality) to ensure alignment with the plan and fast feedback loops.
  • Single-session orchestration that preserves context where needed but prevents cross-task contamination.
  • Use Case: When executing complex implementation plans with multiple independent tasks, dispatch subagents to work in parallel and converge through automated review loops.

Quick Start

Dispatch the plan, extract all tasks, and trigger a programmer subagent to start implementing immediately.

Frequently Asked Questions about subagent-driven-development

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

FAQPage Schema
How do I orchestrate complex implementation plans without context pollution?

You can orchestrate plan execution by dispatching fresh subagents per task within the same session. This approach isolates task handling and prevents cross-task contamination while preserving necessary context.

What is the best way to automate code review during plan execution?

Automating code review during plan execution is best handled through a two-stage review workflow. This process checks spec compliance first, then evaluates code quality to ensure alignment with the implementation plan.

How do I dispatch subagents for parallel task implementation?

You can dispatch subagents for parallel task implementation by triggering a programmer subagent immediately after extracting tasks from your plan. This enables isolated work and fast iteration loops.

Does subagent-driven development work for TDD workflows?

Yes, subagent-driven development supports TDD workflows by using templates to guide subagents through planning, implementation, and automatic reviews. This maintains tight control over quality through structured feedback loops.

Why should I use fresh subagents per task instead of a single session context?

Using fresh subagents per task prevents context pollution and reduces cross-task contamination. It accelerates iteration by isolating task handling while maintaining single-session orchestration for overall plan alignment.