skills.rest
GanyuanRan
GanyuanRanCOMMUNITY
Explore all skills by this author →
GitHub

subagent-driven-development

Orchestrates plan-driven tasks by dispatching subagents and enforcing two-stage reviews.

968|43|Updated Apr 30, 2026
Search Tags:#review#subagent#ai-agent#plan-execution#task-dispatch#workflow-management
One-click install
npx skills add https://github.com/GanyuanRan/Aegis --skill subagent-driven-development-ganyuanran
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/GanyuanRan/Aegis/tree/main/skills/subagent-driven-development
Command: npx skills add https://github.com/GanyuanRan/Aegis --skill subagent-driven-development-ganyuanran

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Orchestrate execution of plan-driven work by dispatching fresh subagents for each independent task and enforcing two-stage reviews.

Core Features & Use Cases

  • Fresh subagent per task: isolate context and reduce cross-task contamination.
  • Two-stage reviews: spec compliance first, then code quality.
  • Structured handoffs: use SubagentContextPacket to share task details and constraints between roles.
  • Checkpoints and drift checks: update progress and verify alignment before moving to the next task.

Quick Start

Dispatch a subagent per independent task from your plan and run sequential two-stage reviews until all tasks are complete.

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 subagents for task dispatch and plan execution?▼

Orchestrate subagents for plan execution by dispatching a fresh subagent for each independent task and enforcing sequential two-stage reviews to ensure spec compliance and code quality.

What is the two-stage review process for AI agent workflow management?▼

The two-stage review process checks spec compliance first, then evaluates code quality, ensuring structured handoffs and reducing cross-task contamination before updating progress checkpoints.

Can I use fresh subagents for intra-session planning with separable tasks?▼

Fresh subagents are best suited for real-time intra-session planning with mostly separable tasks, isolating context per task to reduce contamination and enable per-task implementation loops.

How do subagents report task completion status during plan-driven work?▼

Subagents report task completion status using explicit flags: DONE, DONE_WITH_CONCERNS, BLOCKED, or NEEDS_CONTEXT, facilitating structured handoffs and checkpoint drift checks.

When should I not use a fresh subagent per task for workflow management?▼

Avoid using a fresh subagent per task when work involves tightly coupled, non-separable tasks, as this approach relies on independent task isolation and structured handoffs via SubagentContextPacket.

What is the best way to prevent context contamination across independent AI tasks?▼

The best way to prevent context contamination is dispatching fresh subagents per task, using SubagentContextPacket for structured handoffs, and enforcing checkpoint drift checks between sequential tasks.

Related Skills

Explore More Software Engineering Skills→
saffron-healthsaffron-health
862

generate-spec

Generate phased specification documents from code and documentation analysis.

Official
Advanced
ahostbrahostbr

ls-worker

Bootstrap LiteSuite worker agents with liteharness CLI registration and status reporting.

Community
Intermediate
nmoralescybernmoralescyber

agent-swarm

Decompose complex tasks into parallel subagent workflows with user approval.

Community
Advanced
jaydubya818jaydubya818

subagent-driven-development

Delegate each task to a fresh subagent with two-stage reviews.

Community
Advanced