# Limits & good to know

A short, honest list of what to expect so nothing surprises you.

## Storage and files

- **Visitor uploads:** each file has a size cap, and each project has a total storage allowance.
  If you expect large files (long videos, big datasets), mention it so your AI can plan around the
  limits.
- **Your own images and assets** (logos, photos) share that per-project allowance.
- Your [Dashboard](/dashboard) shows current usage. See [Files & images](/docs/files-and-images)
  for the difference between the two kinds.

## What onvibe is great at

- Web pages and apps that collect, store, and display data.
- Small-to-medium tools, dashboards, forms, galleries, and APIs.
- Anything you want online quickly with a real link to share.

## What it's not built for

- Heavy, large-scale systems (millions of users, huge media libraries, real-time games).
- Long-running background jobs that churn for minutes — apps respond to requests as they come in.
- Replacing a full custom infrastructure for a large business. It's for getting real, useful apps
  live fast.

## One honest caveat: complex front-end frameworks

Most apps "just work" when you ask. A few advanced setups (certain large front-end frameworks that
need a separate build step on your own computer) require your AI to have a local environment to
build in. If you're using a chat-only assistant with no access to your computer, prefer a plain,
straightforward app — your AI will steer you there, and the result is usually simpler anyway. If
you don't know what this means, you almost certainly don't need to worry about it.

## Good habits

- **Save a version before big changes** so you can undo — see [Versions & undo](/docs/versions-and-undo).
- **Use a draft when your app already has real data** — same page.
- **Be specific when something breaks** — see [When something breaks](/docs/troubleshooting).
- **Start small.** It's easier to grow a working app than to debug a giant first attempt.