Page appears CSR-only to crawlers

What this means

The HTML response for your homepage contains fewer than 100 characters of visible text after scripts and styles are removed. Your site is rendering content exclusively in the browser (client-side rendering). AI crawlers — including GPTBot, OAI-SearchBot, and PerplexityBot — do not execute JavaScript. They receive an effectively empty page.

This is the most severe GEO issue you can have. A crawler that sees no content cannot cite you for anything.

How to fix it

The fix depends on your framework. The goal is to return real content in the initial HTML response, before any JavaScript runs.

Next.js

In the App Router, pages are server components by default. The most common cause of accidental CSR-only rendering is placing 'use client' on a root layout or top-level page component.

Move 'use client' down to only the components that require browser APIs or interactivity:

// app/page.tsx — keep this a server component, no 'use client' here
import { HeroSection } from '@/components/HeroSection'
import { InteractiveWidget } from '@/components/InteractiveWidget'

export default function HomePage() {
  return (
    <>
      <HeroSection />
      <InteractiveWidget />
    </>
  )
}

'use client'

export function InteractiveWidget() {
  // browser-only logic lives here, not in the page
}

Nuxt

Check nuxt.config.ts for ssr: false and remove it. SSR is enabled by default in Nuxt:

export default defineNuxtConfig({
  // Remove this line — SSR is on by default
  // ssr: false,
})

SvelteKit

Check src/routes/+layout.ts or individual +page.ts files for export const ssr = false and remove it:

// Remove this line:
// export const ssr = false

SvelteKit renders on the server by default when ssr is not explicitly disabled.

Astro

Astro generates static HTML by default. If you have output: 'server' set and pages that appear blank, ensure your page components return markup in the frontmatter data fetch, not deferred to client scripts:

---
const data = await fetchData() // runs at build or request time, not in the browser
---

<main>
  <h1>{data.title}</h1>
  <p>{data.description}</p>
</main>

Verify the fix

After deploying, fetch the raw HTML and look for paragraph content:

curl -s https://yourdomain.com/ | grep -o '<p[^>]*>[^<]\{20,\}</p>' | head -5

You should see <p> tags containing real text. If the output is empty, the page is still CSR-only. Re-run orino audit to confirm the check passes.