subagent-driven-development

Dispatch fresh subagents for each task and apply two-stage spec and quality reviews.

Updated Mar 27, 2026
One-click install
npx skills add https://github.com/theGobindSingh/cleantank --skill subagent-driven-development-thegobindsingh
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/theGobindSingh/cleantank/tree/main/.agents/skills/subagent-driven-development
Command: npx skills add https://github.com/theGobindSingh/cleantank --skill subagent-driven-development-thegobindsingh

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Subagent-Driven Development orchestrates independent task execution by dispatching dedicated subagents for each task and applying a two-stage review after every task to ensure spec compliance and high code quality.

Core Features & Use Cases

  • Fresh subagent per task to avoid context pollution and enable parallel iteration.
  • Two-stage review process: first verify specification compliance, then assess code quality.
  • Structured task progression with per-task tracking and iterative refinement.

Quick Start

Use Subagent-Driven Development to execute a multi-step plan by dispatching subagents for each task and routing their work through spec and quality reviews.

Frequently Asked Questions about subagent-driven-development

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

FAQPage Schema
How do I automate parallel tasks while preventing context pollution in a single session?

Dispatch a fresh subagent for each task to prevent context pollution and enable fast parallel iteration. This approach maintains minimal context contamination within a single session while executing independent tasks.

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

The two-stage review process first verifies specification compliance, then assesses code quality after every task. This structured workflow ensures spec compliance and high code quality through iterative refinement and per-task tracking.

How do I orchestrate a multi-step plan using subagents?

Orchestrate a multi-step plan by dispatching dedicated subagents for each task and routing their work through spec and quality reviews. This requires dedicated prompts for the implementer, spec-reviewer, and code-quality-reviewer to track task completion.

Does subagent-driven development work for independent tasks within a single session?

Yes, subagent-driven development works well for independent tasks within a single session. It enables fast iteration and minimal context contamination by dispatching fresh subagents and applying a two-stage review for each task.

What are the limitations of using a fresh subagent for each task?

A limitation is that this approach requires dedicated prompts for the implementer, spec-reviewer, and code-quality-reviewer, plus a coordinated workflow to track task completion. It is designed for independent tasks within a single session rather than continuous execution.