subagent-driven-development

Dispatch isolated subagents per task with two-stage spec and quality review gates.

Updated Mar 27, 2026
One-click install
npx skills add https://github.com/CenredJun/openclaw-claudecode-setup-kit --skill subagent-driven-development-cenredjun
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/CenredJun/openclaw-claudecode-setup-kit/tree/main/skills/subagent-driven-development
Command: npx skills add https://github.com/CenredJun/openclaw-claudecode-setup-kit --skill subagent-driven-development-cenredjun

SYSTEM DOCUMENTATION & REQUIREMENTS

## What problem does it solve? The Skill streamlines executing implementation plans by dispatching a fresh, focused subagent for each independent task and enforcing a two-stage review process to ensure spec compliance and high code quality.

## Core Features & Use Cases

  • Isolated execution per task: Assign a fresh implementer subagent to avoid context pollution and enable parallel-safe work.
  • Two-stage review gates: Enforce a spec compliance reviewer first, then a code quality reviewer for every completed implementation.
  • Model selection and status handling: Select models by task complexity and handle implementer statuses like NEEDS_CONTEXT, DONE_WITH_CONCERNS, and BLOCKED with clear remediation steps.
  • Use Case: Run a planned feature rollout by extracting tasks from a plan, dispatching implementer subagents, iterating on reviewer feedback until approvals, and marking tasks complete in your tracker.

### Quick Start Use subagent-driven-development to execute the plan file and dispatch an implementer subagent for the first task, answering any questions before implementation.

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 implementation plans using isolated subagents?

Isolated subagent execution dispatches a fresh implementer subagent for each independent task to prevent context pollution and enable parallel-safe development. This approach streamlines planned feature rollouts by maintaining focused execution contexts throughout the workflow.

What is a two-stage review process for code implementation?

A two-stage review process enforces a spec compliance reviewer first, followed by a code quality reviewer for every completed implementation. This ensures both specification adherence and high code quality standards before tasks are marked complete in your tracker.

How do I handle blocked or incomplete subagent task statuses during implementation?

Handling implementer statuses like NEEDS_CONTEXT, DONE_WITH_CONCERNS, and BLOCKED requires applying clear remediation steps based on the specific status returned. The workflow iterates on reviewer feedback until approvals are met before marking tasks complete.

Can I select different models based on task complexity in a subagent workflow?

Model selection by task complexity is supported within the subagent workflow, allowing you to assign appropriate model capabilities to each independent task. This optimizes execution while maintaining the two-stage review gates for spec compliance and quality.

Does subagent-driven development integrate with git and task tracking systems?

Subagent-driven development integrates with git workflows and task tracking systems to mark completed tasks and manage code review gates. This integration ensures implementation plans execute with proper version control and progress tracking throughout the development cycle.