subagent-driven-development

Dispatch fresh subagents per task with two-stage reviews in a single session.

Updated Feb 23, 2026

Search Tags:#automation #workflow #tdd #subagent #code-review #plan-execution #task-dispatch

Ready to run

npx skills add https://github.com/rishikanthc/ml-superpowers --skill subagent-driven-development-rishikanthc

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/rishikanthc/ml-superpowers/tree/main/skills/subagent-driven-development
Command: npx skills add https://github.com/rishikanthc/ml-superpowers --skill subagent-driven-development-rishikanthc

SYSTEM DOCUMENTATION & REQUIREMENTS

## What problem does it solve? Execute implementation plans by dispatching fresh subagents per task, with a two-stage review process (spec compliance followed by code quality) to accelerate iteration while preserving quality.

## Core Features & Use Cases

Fresh per-task subagents with independent execution to prevent context pollution
Two-stage reviews per task: spec compliance first, then code quality
Single-session execution with clear task progression and automatic completion
Suitable for parallel or sequential task workflows across complex plans

### Quick Start Dispatch a fresh subagent for each task in your plan and apply the two-stage review process to ensure 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 implementation plans without context pollution across independent tasks?▼

Execute implementation plans by dispatching fresh subagents per task. This per-task isolation prevents context pollution, allowing independent execution within a single session while accelerating fast iteration.

What is the best way to automate code review during plan execution workflows?▼

Automate code review by applying a two-stage review process per task: spec compliance first, then code quality. This ensures self-reviews are integrated directly into the plan execution workflow before marking tasks complete.

Why does subagent task dispatch use a two-stage review process?▼

Subagent task dispatch uses a two-stage review process to separate spec compliance from code quality. This isolates functional requirements verification from structural code assessment, preserving quality during fast iteration.

Can I use subagent orchestration for sequential task workflows across complex plans?▼

Yes, subagent orchestration supports both parallel and sequential task workflows across complex plans. It dispatches fresh subagents for independent tasks within a single session to maintain clear task progression and automatic completion.

When should I not use per-task subagents for workflow automation?▼

Avoid per-task subagents when tasks are heavily interdependent rather than largely independent. The workflow is designed for isolated execution and fast iteration, so tightly coupled tasks may break the single-session progression model.