EUmove — Demo

A single-page, client-only React prototype of EUmove, an AI-assisted cross-border migration companion for EU citizens. This demo follows a sample user (Sophie van der Berg) moving from the Netherlands to Germany, and shows personalised tasks, opportunities, a profile, and an assistant chat.

The whole app lives inside index.html:

• React 18 + Babel Standalone via CDN (no build step)
• Fraunces + Inter web fonts
• All state is in-memory (ticks, notes, and profile edits are lost on refresh — by design, no backend)

Run locally

Just open index.html in a browser. Or serve it with any static file server:

npx serve .
# or
python3 -m http.server 8000

Then visit http://localhost:8000.

Deploy to Vercel (free)

Option A — Vercel CLI (fastest)

npm i -g vercel
vercel            # first deploy (preview)
vercel --prod     # production deploy

Accept the defaults. Vercel auto-detects it as a static site; no framework preset needed.

Option B — GitHub + Vercel dashboard

1. Create a new repository on GitHub and push this folder:

   git remote add origin https://github.com/<your-user>/eumove-demo.git
   git push -u origin main

2. Go to vercel.com/new and import the repo.
3. Leave all build settings empty (it's a pure static site).
4. Click Deploy. Vercel gives you a free *.vercel.app URL.

Subsequent pushes to main auto-deploy.

Option C — Drag & drop

Go to vercel.com/new, pick "Deploy a template / upload a folder", and drop this folder.

What's inside

.
├── index.html       # entire app (React via CDN)
├── vercel.json      # static-site routing config
├── .gitignore
└── README.md

Notes / known limitations

This is a demo, so:

• No persistence — refreshing resets completed tasks, notes, and profile edits.
• The Profile "Save changes" button is cosmetic; fields already update live.
• The assistant has 4 canned answers plus a generic fallback (see the AI object in index.html).
• departureDate and daysLeft aren't linked; the "19 days" count is hard-coded.

If this prototype graduates to a real product, the natural next steps would be: a small backend or edge KV for persistence, real EUDIW integration, and a proper markdown renderer for assistant answers.