skills.rest
gujiachun-rainbow
gujiachun-rainbowCOMMUNITY
Explore all skills by this author →
GitHub

Subagent-Driven Development

Dispatch fresh subagents per task with automated reviews between tasks.

8|1|Updated Apr 15, 2026
Search Tags:#automation#subagent#software-development#code-review#task-orchestration#plan-execution
One-click install
npx skills add https://github.com/gujiachun-rainbow/RainClaw --skill subagent-driven-development-gujiachun-rainbow
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: Subagent-Driven Development
Source: https://github.com/gujiachun-rainbow/RainClaw/tree/main/.trae/skills/subagent-driven-development
Command: npx skills add https://github.com/gujiachun-rainbow/RainClaw --skill subagent-driven-development-gujiachun-rainbow

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Orchestrates complex development plans by delegating each task to a fresh subagent, ensuring isolated context and rapid iteration with built-in reviews.

Core Features & Use Cases

  • Fresh subagent per task to prevent context leakage and enable parallel progress
  • Automatic code reviews between tasks to catch issues early
  • End-to-end task orchestration with plan loading, execution, review, and final verification
  • Use cases: implementing multi-step features, performing plan-driven development, ensuring quality gates in automation

Quick Start

Load a plan and start executing tasks with a new subagent for each task, with review between tasks.

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 multi-step software development plans without context leakage?▼

Subagent-driven development orchestrates complex plans by dispatching a fresh subagent for each task. This provides isolated context for rapid iteration and uses automatic review gates between tasks to ensure quality.

What is the best way to automate code reviews between sequential development tasks?▼

Automating code reviews between tasks is best handled by dispatching fresh subagents to execute an implementation plan. This creates review gates that catch issues early and provide git-style feedback loops.

How do I execute an implementation plan with automated per-task testing and reviews?▼

You execute an implementation plan by loading it into the session and starting task execution. A new subagent is dispatched for each task, applying per-task testing and automated reviews to ensure traceability.

Can I use task orchestration for plan-driven development within a single session?▼

Yes, task orchestration applies to tasks within the same session. It satisfies end-to-end orchestration by loading the plan, executing tasks with fresh subagents, and running final verification.

Why does automated plan execution need fresh subagents for each task?▼

Automated plan execution needs fresh subagents to prevent context leakage between tasks. This isolation enables parallel progress and ensures that automated reviews evaluate each task independently.

Related Skills

Explore More Software Engineering Skills→
darthlinuxerdarthlinuxer
1

subagent-driven-development

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

Community
Advanced
ramonclaudioramonclaudio
4

end

Archive Claude session state and health checks to the .handoff directory.

Community
Advanced
SighthesiaSighthesia
4

subagent-driven-development

Dispatch a fresh subagent per task and review code between tasks.

Community
Advanced
yanochka11yanochka11

subagent-driven-development

Delegate software development tasks to subagents with validation workflows.

Community
Advanced