간단한 데이터 저장

Most apps need to remember things: your to-do list, your saved notes, the dark-mode setting you chose. Without storage, everything disappears when you refresh the page.

The good news: you can go from 'remembers nothing' to 'remembers everything' just by describing what to save — AI handles the storage code.

There are levels of storage, from simplest to most powerful. Start low and climb only when you need to:

• In memory — fast but forgotten on refresh
• Local storage — saved in the user's browser, survives refresh
• A database — shared across devices and users, the real deal

Telling AI which rung you want keeps your app as simple as possible.

The easiest persistent storage is localStorage — a little notebook the browser keeps for your app. It's perfect for settings and small lists on a single device.

You don't need a server or account for this. A prompt like the one below adds it instantly.

localStorage stores text by a name (a 'key'). To save objects or lists, you turn them into JSON text first, then turn them back when reading.

Run this to see the pattern AI uses: save a value, then read it back.

localStorage only lives in one browser on one device. If you want users to see their data on their phone and laptop, or share data between users, you need a database on a server.

That sounds scary, but tools make it easy. You'll describe what to store, and AI sets up the database calls.

You don't need to run your own server. Hosted databases give you storage with almost no setup:

• Supabase — a friendly database with built-in login, great with AI tools
• Firebase — Google's real-time database
• Airtable — a spreadsheet that acts like a database

Builders like Bolt, Lovable and Replit can connect these for you when you ask.

To add real storage, describe what you're saving and where. Naming Supabase (a popular choice) tells AI exactly which setup to generate.

This prompt turns a local-only app into one with a shared, persistent database.

A database stores data in tables — like spreadsheets with named columns. A 'notes' table might have a title, content, and a date. Each saved note is one row.

You don't need to master database design. Just tell AI what fields each thing has, and it builds the table for you.

Almost every data app does four things, nicknamed CRUD: Create new items, Read them back, Update them, and Delete them. If your app can do all four, it's a real tool.

You can ask AI to add them one at a time, testing each before moving on.

The pro move is to start at the lowest rung that works. Building a quick prototype? localStorage. Need it on every device? Move up to a database. AI can even migrate you — 'switch my app from localStorage to Supabase.'

Don't over-engineer early. Match the storage to what your app actually needs today.

Data is precious — users get upset if it vanishes. Hosted databases like Supabase and Firebase keep backups for you, which is another reason to graduate from localStorage for anything important.

You can also let users export their own data as a safety net. Ask AI to add a one-click backup so nobody loses their work.

You built a notes app that uses the browser's localStorage. A user complains they can't see their notes when they open the app on their phone. What kind of storage do they actually need?

You learned the storage ladder: in-memory, then localStorage for single-device persistence, then a real database for cross-device, shared data.

You saw how to save settings in the browser, when to upgrade to hosted databases like Supabase, how data lives in tables, and the CRUD actions every data app needs. Start simple and let AI upgrade you. Next: keeping your API keys and secrets safe.