skills.rest
bakwankawa
bakwankawaCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

subagent-driven-development

Orchestrates plan-driven tasks with subagents and two-stage reviews in Cursor.

1|Updated Feb 1, 2026
Search Tags:#workflow#subagents#code-review#software-engineering#Cursor#plan-execution#two-stage-review
One-click install
npx skills add https://github.com/bakwankawa/lazy-code-skill --skill subagent-driven-development-bakwankawa
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/bakwankawa/lazy-code-skill/tree/main/.cursor/skills/superpowers/subagent-driven-development
Command: npx skills add https://github.com/bakwankawa/lazy-code-skill --skill subagent-driven-development-bakwankawa

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill orchestrates plan-driven work by dispatching fresh subagents to handle each task within the same Cursor session, enforcing a two-stage review process to improve quality and speed.

Core Features & Use Cases

  • Fresh subagent per task to isolate context and reduce cross-task pollution.
  • Two-stage review: spec compliance first, then code quality review, enabling rapid yet thorough iteration.
  • In-session orchestration via the Subagents feature with prompts for implementers and reviewers to drive execution.

Quick Start

Use the subagent-driven development workflow to execute a plan by dispatching implementer and reviewer subagents per task. See the SKILL's prompt templates (implementer-prompt.md, spec-reviewer-prompt.md, code-quality-reviewer-prompt.md) for the exact prompts.

Frequently Asked Questions about subagent-driven-development

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

FAQPage Schema
How do I isolate context and reduce cross-task pollution when executing plans with subagents?▼

To isolate context and reduce cross-task pollution when executing plans with subagents, dispatch a fresh subagent for each task within your Cursor session to ensure independent execution and cleaner state management.

What is the two-stage review process for subagent-driven development?▼

The two-stage review process for subagent-driven development sequentially checks spec compliance first, then evaluates code quality, enabling rapid yet thorough iteration and improved output quality for each dispatched task.

How do I orchestrate plan execution using implementer and reviewer subagents in Cursor?▼

To orchestrate plan execution using implementer and reviewer subagents in Cursor, use the pre-defined prompt templates for implementers, spec reviewers, and code quality reviewers to drive in-session task execution and reviews.

Does subagent-driven development work outside of Cursor sessions?▼

Subagent-driven development is designed to work within Cursor sessions and relies on the platform's Subagents feature to orchestrate plan execution, dispatch tasks, and enforce the two-stage review process.

Why use fresh subagents per task instead of a single agent for plan execution?▼

Using fresh subagents per task isolates context, reduces cross-task pollution, and enforces a structured two-stage review, which improves iteration speed and output quality compared to a single agent handling the entire plan.

Related Skills

Explore More Software Engineering Skills→
tuantiensiutuantiensiu
1

subagent-driven-development

Dispatch fresh subagents per task with mandatory inter-task code reviews.

Community
Advanced
mguilarduccimguilarducci
1

execute

placeholder

Community
Advanced
kjayleekjaylee

subagent-dev

Dispatch implementation tasks to subagents with two-stage spec and code review.

Community
Advanced
tusosostusosos

subagent-driven-development

Dispatch fresh subagents per task to isolate context and run two-stage reviews.

Community
Advanced