gk-new-app

Generate CLAUDE.md and gk/ directory scaffolding for greenfield projects.

1|Updated Jul 4, 2026
One-click install
npx skills add https://github.com/gkganesh12/gk-stack --skill gk-new-app
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: gk-new-app
Source: https://github.com/gkganesh12/gk-stack/tree/main/skills-extra/gk-new-app
Command: npx skills add https://github.com/gkganesh12/gk-stack --skill gk-new-app

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This skill solves the problem of inconsistent project scaffolding and context loss in AI-assisted development by enforcing a rigorous, spec-first architecture that ensures the AI agent always has a clear, verifiable plan.

Core Features & Use Cases

  • Automated Project Scaffolding: Generates a standardized gk/ directory structure and CLAUDE.md to ensure the agent has immediate access to architecture, specs, and progress tracking.
  • Spec-Driven Development: Enforces the creation of SPEC.md and PLAN.md before any code is written, preventing drift and ensuring incremental, verifiable progress.
  • Use Case: When starting a new greenfield project, use this skill to instantly generate the entire project kit, ensuring that every feature is planned, sequenced by dependency, and ready for the agent to execute safely.

Quick Start

Run the gk-new-app skill to initialize the project structure and generate the required specification and planning documents for your new application.

Frequently Asked Questions about gk-new-app

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

FAQPage Schema
How do I set up spec-driven scaffolding for AI-assisted development?

Spec-driven scaffolding for AI-assisted development is set up by generating a standardized directory structure and a CLAUDE.md file to provide the AI agent with immediate architecture context and progress tracking.

What's the best way to prevent context loss when initializing a greenfield project with AI?

Preventing context loss in a greenfield AI project is achieved by enforcing a rigorous, spec-first architecture that requires creating SPEC.md and PLAN.md documents before any code is written.

Why does my AI coding agent lose architectural consistency across project sessions?

Architectural consistency is lost when the agent lacks standardized documentation, which is solved by generating grounding artifacts that ensure dependency-aware task planning and verifiable progress.

Can I use automated project scaffolding for dependency-aware task planning?

Automated project scaffolding supports dependency-aware task planning by generating a project kit where every feature is planned, sequenced by dependency, and ready for the agent to execute safely.

Do I need to create SPEC.md before writing code in a spec-driven workflow?

Creating SPEC.md before writing code is required in a spec-driven workflow to prevent drift and ensure incremental, verifiable progress throughout the development lifecycle.

When should I not use spec-driven project initialization?

Spec-driven project initialization should not be used when your project already has an established directory structure and existing documentation, as it targets greenfield project initialization.