Back to all days

Day 26 — Digital Strategy

Tech Stack Decision Engine

“Should we use Next.js or Remix?” is never actually about Next.js or Remix.

An MCP server that weighs team size, budget, timeline, and goals — then recommends a stack with receipts. 32 technologies scored across 12 criteria, with compatibility data and real-world project analysis.

Example Queries

“Should we use Next.js or Remix for our new SaaS product?”
“Recommend a stack for a 3-person team shipping an MVP in 8 weeks”
“How well do Supabase, Next.js, and Clerk work together?”
“What are the tradeoffs between PostgreSQL and MongoDB for e-commerce?”
“Show me all frontend frameworks ranked by developer experience”
“Analyze the HealthBridge enterprise project — what stack would you pick?”

Tools

6 MCP Tools

ToolWhat It Does
evaluate_technologyDeep evaluation of a specific technology — scores across 12 criteria, strengths, weaknesses, and best-fit pairings.
compare_technologiesHead-to-head comparison of 2–5 technologies — weighted scores, per-criteria winners, and an overall recommendation.
recommend_stackFull-stack recommendation based on team size, budget, timeline, and priority — returns the best pick for each layer.
check_compatibilityCheck how well technologies work together — integration scores, friction points, and missing stack layers.
analyze_projectAnalyze sample projects with real constraints — see recommended stacks, rationale, and tradeoffs for FinTech, Healthcare, E-Commerce, and more.
tech_searchSearch and filter technologies by category, maturity, learning curve, license, or GitHub stars.

The Engine

How Decisions Get Made

Every technology is scored against 12 weighted criteria. Your constraints — team size, budget, timeline — adjust those weights to surface the right answer for your situation:

Score

32 technologies rated across 12 criteria — performance, scalability, DX, type safety, learning curve, hiring pool, community, docs, maintenance cost, time to market, hosting cost, and vendor lock-in.

Weight

Your constraints shift the weights. Small team? Learning curve and DX matter more. Enterprise? Scalability and type safety get boosted.

Match

Technologies are ranked per-category with compatibility checks — ensuring your frontend, backend, database, and hosting all play well together.

Decide

Get a full-stack recommendation with rationale and tradeoffs. Compare alternatives side by side. Ship with confidence.

Why This Build

Why a Tech Stack Decision Engine?

Every startup and every agency has the same meeting: “What should we build this on?” The answer usually comes from whoever's loudest, not from data. Frameworks get picked because the CTO used them at their last job, not because they fit the constraints.

The real question is never “which framework is best” — it's “which framework is best for this team, this budget, this timeline, and this goal?” A 3-person team shipping in 8 weeks needs a different stack than a 15-person team building for HIPAA compliance.

This tool makes those tradeoffs explicit. Ask a question about your stack, get an answer backed by scores, compatibility data, and real-world project analysis. Second Digital Strategy build of the challenge.

Demo Data

What's Inside

32

Technologies

6

Categories

12

Criteria

384

Scores

4

Projects

22

Recommendations

32 technologies across frontend (Next.js, Remix, SvelteKit, Astro, Nuxt, Angular), backend (Express, Fastify, NestJS, Django, FastAPI, Rails), databases (PostgreSQL, MySQL, MongoDB, Supabase, PlanetScale, Redis), hosting (Vercel, AWS, Fly.io, Railway, Render, Cloudflare Workers), auth (Clerk, Auth.js, Supabase Auth, Firebase Auth), and styling (Tailwind, CSS Modules, Styled Components, shadcn/ui). 4 sample projects from FinTech MVP to enterprise healthcare with full stack recommendations.

Setup

Try It Locally

cd day26-tech-stack
npm install && npm run seed && npm run build

Then add the config to Claude Desktop and restart.

Need help choosing the right stack for your project?

If your team is stuck in a framework debate instead of shipping, let's turn that conversation into a decision.