skills.rest
luicabref97
luicabref97COMMUNITY
Explore all skills by this author →
GitHub

sadd:subagent-driven-development

Dispatch fresh subagents for isolated implementation tasks with code-review checkpoints.

Updated Apr 6, 2026
Search Tags:#tdd#subagent#code-review#task-orchestration#parallel-execution#subagent-driven-development
One-click install
npx skills add https://github.com/luicabref97/sushi-jungle-web --skill sadd-subagent-driven-development-luicabref97
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: sadd:subagent-driven-development
Source: https://github.com/luicabref97/sushi-jungle-web/tree/main/.agents/skills/sadd-subagent-driven-development
Command: npx skills add https://github.com/luicabref97/sushi-jungle-web --skill sadd-subagent-driven-development-luicabref97

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill addresses slow or error-prone implementation workflows by dispatching isolated subagents per task and enforcing code-review checkpoints so changes are high-quality and iteration is fast.

Core Features & Use Cases

  • Per-task isolation: Launch a fresh subagent for each task or issue to avoid context pollution and accidental cross-task interference.
  • Integrated code review: Run a dedicated code-review subagent after implementation to catch Critical and Important issues before progressing.
  • Sequential and parallel modes: Execute tightly coupled tasks sequentially or run independent investigations or fixes in parallel with batch reviews.
  • Use case: Break a large development plan into discrete tasks, have each implemented and tested by a separate agent, run automated code reviews, apply fixes, and finish with a final validation before merging.

Quick Start

Use the skill to run a plan with one fresh implementation subagent per task and perform code review after each task.

Frequently Asked Questions about sadd:subagent-driven-development

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

FAQPage Schema
How do I isolate independent tasks for parallel execution using subagents?▼

To isolate tasks for parallel execution, dispatch a fresh subagent for each independent implementation task. This per-task isolation prevents context pollution and cross-task interference, enabling fast parallel execution of unrelated investigations or fixes.

How do I automate code review checkpoints during development?▼

Automate code review checkpoints by running a dedicated review subagent after implementation. This mechanism catches Critical and Important issues before progressing, enforcing quality control through integrated follow-up fix cycles and final validation.

What is subagent-driven development and when should I use it?▼

Subagent-driven development dispatches isolated subagents per task with enforced code-review checkpoints. Use it to break large development plans into discrete tasks, run batch investigations of unrelated failures, or execute scenarios requiring isolated execution contexts.

Can I execute tightly coupled tasks sequentially with subagents?▼

Yes, you can execute tightly coupled tasks sequentially by dispatching fresh implementation subagents. The workflow supports both sequential execution for coupled tasks and parallel execution for independent investigations.

How do I prevent context pollution when running multiple coding tasks?▼

Prevent context pollution by launching a fresh subagent for each task or issue. Per-task isolation ensures each implementation runs in an isolated execution context, avoiding accidental cross-task interference and maintaining development speed.

Related Skills

Explore More Software Engineering Skills→
databrickslabsdatabrickslabs
38

subagent-driven-development

Execute multi-task implementation plans with fresh subagents and two-stage reviews.

Official
Advanced
haleyrchaleyrc
2

feature-merge

Squash merge feature branch commits into main with a Conventional Commit message.

Community
Intermediate
automagik-devautomagik-dev
328

fix

Automate code gap resolution with iterative fixes and escalation.

Official
Intermediate
stabilefrisurstabilefrisur

subagent-driven-development

Dispatch fresh subagents per task with two-stage spec and quality reviews.

Community
Advanced