subagent-driven-development

Executes implementation plans by dispatching subagents with two-stage review checkpoints.

Updated May 26, 2026
One-click install
npx skills add https://github.com/anukkrit149/anukkrit-skills --skill subagent-driven-development-anukkrit149
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: subagent-driven-development
Source: https://github.com/anukkrit149/anukkrit-skills/tree/main/cloud/skills/subagent-driven-development
Command: npx skills add https://github.com/anukkrit149/anukkrit-skills --skill subagent-driven-development-anukkrit149

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Subagent-driven development prevents slow, error-prone implementation by breaking an execution plan into independent tasks handled by fresh subagents with checkpoints for correctness and quality.

Core Features & Use Cases

  • Per-task dispatch: Runs a dedicated implementer subagent for each task to avoid cross-task context pollution and keep iterations fast.
  • Two-stage review gate: After each task, runs a spec compliance review first and a code quality review second, looping until both pass.
  • Interactive task execution: Allows the implementer subagent to ask questions before starting if requirements or acceptance criteria are unclear.
  • Use in session: Best for executing an implementation plan where tasks are mostly independent within the same session.

Quick Start

Use subagent-driven-development to execute an existing implementation plan by extracting each task, dispatching an implementer per task, and running spec then code quality reviews after each task until approved.

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 an implementation plan with independent tasks while maintaining code quality?

Subagent-driven development executes an implementation plan by dispatching a fresh implementer subagent per task, followed by spec compliance and code quality review checkpoints to ensure correctness and maintainable code.

What is two-stage review in task orchestration and how does it work?

Two-stage review runs a spec compliance review first and a code quality review second after each task implementation, looping iterative fixes and re-reviews until both stages pass before moving forward.

How do I prevent cross-task context pollution during test-driven development?

Dispatching a dedicated fresh subagent for each independent task prevents cross-task context pollution, keeping iterations fast and isolated while executing an implementation plan within the same session.

When should I use per-task subagent dispatch for implementation?

Use per-task subagent dispatch when tasks in your implementation plan are mostly independent and require high correctness, benefiting from iterative spec compliance and code quality review loops.

Can the implementer subagent ask questions before starting a task?

Yes, interactive task execution allows the implementer subagent to ask questions before starting if requirements or acceptance criteria for the task are unclear.

Does subagent-driven development include a final code review for the whole implementation?

Yes, after all independent tasks pass their two-stage review checkpoints, a final code review runs for the whole implementation to ensure overall spec compliance and code quality.