# Overview

Source: https://docs.gryt.chat/docs/site

The gryt.chat marketing site and blog — Vite, React, and MDX

The Gryt site (`packages/site`) is the public-facing website at [gryt.chat](https://gryt.chat). It serves the landing page, feature overview, and blog.

## Stack

- **Vite** for bundling and dev server
- **React 19** with React Router for SPA routing
- **MDX** for blog posts (via `@mdx-js/rollup` with remark plugins)
- **Radix UI** for accessible primitives (Dialog, Navigation Menu, Tooltip)
- **CSS Modules** for scoped styling with CSS custom properties

## Project structure

```
site/
├── content/
│   ├── blog/           # MDX blog posts
│   └── changelog/      # MDX release notes, one file per version
├── public/             # Static assets (images, videos, logo)
├── scripts/            # Image optimisation, prerendering, meta checks
├── src/
│   ├── components/     # Navbar, Footer, Hero, Download, Lightbox, etc.
│   ├── lib/
│   │   ├── blog.ts     # Blog loader (Vite glob import)
│   │   ├── changelog.ts
│   │   └── title.ts
│   ├── pages/          # Route components
│   ├── styles/
│   ├── App.tsx         # The route table
│   └── index.css       # Global styles and CSS variables
└── vite.config.ts
```

## Routes

| Path | Page |
|------|------|
| `/` | The landing page, including the download section |
| `/why-gryt` | The longer argument |
| `/blog`, `/blog/:slug` | Blog index and posts |
| `/changelog`, `/changelog/:version` | Release notes, one MDX file per version |
| `/privacy`, `/terms`, `/community-guidelines` | The policy pages. Each also answers a longer alias, `/privacy-policy`, `/terms-of-use` and `/guidelines` |
| `/invite` | Where an invite link lands |
| `/auth/callback` | The OIDC redirect target |
| `*` | Not found |

## Getting started

```bash
cd packages/site
npm install
npm run dev
```

Open [http://localhost:5173](http://localhost:5173).

## Build

```bash
npm run build     # production bundle → dist/
npm run preview   # preview the production build locally
```

`build` is more than a bundle. In order it typechecks with `tsc -b`, optimises
images, runs the Vite build, prerenders the blog and changelog so each post has
real HTML for crawlers and link unfurls, and finishes with `check-meta`, which
fails the build if any page is missing a title suffix, an `og:image` or a
consistent canonical.