subagent-driven-development

Dispatch fresh subagents per task and enforce spec compliance before code quality review.

Updated Feb 16, 2026
One-click install
npx skills add https://github.com/wewetv1987-cell/openclaw-backup --skill subagent-driven-development-wewetv1987-cell
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/wewetv1987-cell/openclaw-backup/tree/main/skills/clawgator-superpowers/skills/subagent-driven-development
Command: npx skills add https://github.com/wewetv1987-cell/openclaw-backup --skill subagent-driven-development-wewetv1987-cell

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you execute implementation plans efficiently when the work is split into mostly independent tasks, while keeping quality high through structured reviews and fresh context for each step.

Core Features & Use Cases

  • Per-Task Subagent Dispatch: Sends a new implementer subagent for each task so work stays isolated and context pollution is avoided.
  • Two-Stage Review Gate: Reviews every task first for spec compliance and then for code quality before marking it complete.
  • Plan-to-Execution Workflow: Supports reading a plan once, extracting tasks, tracking progress, handling questions, and finishing the branch cleanly.
  • Use Case: Ideal for a session where you already have a task plan and want each item implemented, checked, fixed if needed, and approved without switching to a separate parallel workflow.

Quick Start

Tell me to execute your implementation plan with subagent-driven development, and I will break it into tasks, dispatch fresh subagents, and review each task for spec compliance and code quality.

Frequently Asked Questions about subagent-driven-development

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

FAQPage Schema
How do I execute task plans with subagents while keeping context clean?

Subagent-driven development orchestrates task execution by dispatching a fresh implementer subagent for each independent task, ensuring context stays isolated and progress continues without human handoff between tasks.

What is the two-stage review process for subagent task execution?

The two-stage review gate evaluates every completed task first for spec compliance and then for code quality before marking it complete, ensuring implementations meet requirements and maintain standards.

How do I track implementation progress when running multiple subagent tasks?

Use TodoWrite tracking to monitor progress across extracted tasks during same-session execution, allowing you to handle questions and manage the workflow until final branch completion.

Can I use subagent-driven development for dependent tasks in the same session?

This approach applies to same-session execution where tasks are mostly independent. If your tasks have heavy dependencies requiring shared context, the fresh subagent isolation model may not fit.

What's the best way to automate plan-to-execution workflows without context pollution?

Read the implementation plan once, extract tasks, dispatch fresh subagents per task, run spec and code quality reviews, and complete the branch cleanly without switching to a separate parallel workflow.