stack-architect

Generates tech stack decisions covering backend necessity, API selection, and scale-matched deployment plans.

Updated Apr 5, 2026
One-click install
npx skills add https://github.com/Simon-YHKim/eject-button --skill stack-architect-simon-yhkim
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: stack-architect
Source: https://github.com/Simon-YHKim/eject-button/tree/main/.claude/skills/stack-architect
Command: npx skills add https://github.com/Simon-YHKim/eject-button --skill stack-architect-simon-yhkim

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Choosing the right technical architecture is hard: teams often over-engineer MVPs or under-provision growing services. This Skill turns vague questions like "do I need a backend?" or "what API should I use?" into a concrete, scale-appropriate architecture decision. ## Core Features & Use Cases - Backend Necessity Decision Tree: Determines whether a service needs a backend, a BaaS (Supabase/Firebase), or frontend-only based on data sharing, auth, payments, AI inference, and realtime requirements. - API Selection Matrix: Compares REST, GraphQL, tRPC, gRPC, and BaaS-direct approaches against project type and team shape. - Scale-Matched Deployment & Stability Roadmap: Maps user scale (MVP to 1M+) to platforms and monthly cost estimates, plus a four-phase stability roadmap from manual deploys to multi-AZ with SLA targets. - Use Case: A solo founder asks "백엔드 필요해? 규모에 맞는 배포 알려줘" and receives an ARCHITECTURE.md with a chosen stack, deployment platform, cost estimate, and scaling trigger metrics. ## Quick Start Ask the assistant to decide the tech stack for your service idea, including whether a backend is needed, which API style fits, and a deployment plan matched to your expected user scale.

Frequently Asked Questions about stack-architect

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

FAQPage Schema
How do I decide if my app needs a backend?

Check whether your service involves user-to-user data sharing, authentication, payments, AI inference, or realtime features like chat. If any apply, you need a backend or BaaS; purely static tools and calculators can run frontend-only.

REST vs GraphQL vs tRPC: which API should I choose?

REST fits simple CRUD apps like blogs and e-commerce, GraphQL suits complex relational data and mobile optimization, and tRPC works best for TypeScript fullstack projects with small teams. gRPC is reserved for high-performance server-to-server communication.

What deployment platform fits an MVP with under 100 users?

For MVP scale, Vercel plus Supabase Free tier is recommended at roughly zero monthly cost. As you grow toward 1K users, move to Cloudflare with Supabase Pro at around $25-50 per month.

When should I use Supabase or Firebase instead of a custom backend?

Use a BaaS when you only need personal data storage, auth, or simple CRUD without complex server logic. It suits MVPs and small-scale apps; custom servers become necessary for realtime systems, AI inference, or microservice architectures.

What are the limitations of a frontend-only architecture?

Frontend-only apps cannot support user-to-user data sharing, secure payments, server-side AI inference, or realtime features like chat and notifications. If any of these requirements emerge, you must add a backend or BaaS layer.