subagent-driven-development

Delegate tasks to isolated subagents with two-stage spec and code quality reviews.

Updated Mar 7, 2026
One-click install
npx skills add https://github.com/randoneering/nix-flake-mirror --skill subagent-driven-development-randoneering
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/randoneering/nix-flake-mirror/tree/main/home/programs/opencode/skills/subagent-driven-development
Command: npx skills add https://github.com/randoneering/nix-flake-mirror --skill subagent-driven-development-randoneering

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates and coordinates complex task execution by delegating work to isolated subagents, preventing context leakage and enabling rapid, structured reviews across multiple tasks.

Core Features & Use Cases

  • Fresh subagent per task with isolated context to avoid cross-task contamination
  • Two-stage review: spec compliance first, then code quality
  • Self-review by implementer and escalation flow for blocked or unclear tasks
  • Predefined templates for implementer, spec reviewer, and code quality reviewer
  • Ideal for plan-based development workflows, iterative task execution, and multi-task coordination

Quick Start

Dispatch the first implementer subagent for Task 1 and follow the two-stage review workflow to complete 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 delegate code implementation tasks to subagents without context leakage?

You can delegate code implementation tasks to fresh subagents with isolated contexts to prevent cross-task contamination. This approach uses predefined templates for implementers and structured escalation flows for blocked tasks.

What is the best way to automate two-stage code reviews for plan-based development?

Automating two-stage code reviews involves dispatching a spec reviewer followed by a code quality reviewer. This structured review workflow ensures both specification compliance and code quality are checked sequentially.

How does task decomposition work with isolated subagents in iterative development?

Task decomposition with isolated subagents works by orchestrating task execution through fresh agents per task. This preserves context and enables rapid, structured reviews across multiple tasks in iterative development.

Do I need predefined templates to coordinate tasks with focused subagents?

Predefined templates are required to coordinate tasks with focused subagents. They provide the necessary prompts for the implementer, spec reviewer, and code quality reviewer roles to execute structured workflows.

Can I use subagent-driven development for multi-task coordination and iterative execution?

Subagent-driven development is ideal for multi-task coordination and iterative execution. It orchestrates complex workflows by delegating work to isolated subagents, enabling rapid and structured reviews across tasks.

Why does my subagent context leak between tasks during plan execution?

Subagent context leaks between tasks during plan execution when fresh agents are not used per task. Using isolated contexts for each task prevents cross-task contamination and ensures structured reviews.