URL: /rules/ax/content-without-js

---
title: "Content Without JavaScript"
description: "Flags main content that only appears after JavaScript runs — invisible to agents that read raw HTML"
---

Compares the content in your **raw HTML** against the **JavaScript-rendered DOM** and flags pages where significant main content only exists after JS runs. Many AI agents and crawlers read raw HTML without executing JavaScript, so anything injected client-side is invisible to them.

| | |
|---|---|
| **Rule ID** | `ax/content-without-js` |
| **Category** | [Agent Experience](/rules/ax) |
| **Scope** | Per page |
| **Severity** | info |
| **Weight** | 2/10 |

<Note>This rule is a **recommendation** — it never penalizes your score. It is **cloud + render-backed**: it only produces a result on a cloud audit with browser rendering enabled. CLI-only and non-rendered audits skip it gracefully (no error, no finding).</Note>

## What it checks

The rule needs both versions of the page:

- **Raw HTML** — the initial server response, before any JavaScript runs (what a no-JS agent sees).
- **Rendered DOM** — the page after the cloud `render` service executes JavaScript in a real browser.

It extracts the visible text from each with the same pipeline and compares word counts. A page is flagged (`info`) only when **all** of the following hold:

- the rendered page has at least **200 words** of content,
- at least **100 words** exist *only* after JavaScript runs, and
- the raw HTML covers **60% or less** of the rendered content.

Server-rendered (SSR/SSG) pages — where the raw HTML already carries the content — pass. Single-page-app shells that ship a near-empty `<div id="root">` and hydrate everything client-side are flagged.

## Why it matters

A growing share of traffic comes from agents that fetch raw HTML and never run JavaScript: LLM crawlers, answer engines, link unfurlers, and many automation tools. Content that only appears post-hydration is invisible to all of them — your page may look complete in a browser yet read as empty to an agent. Putting the core content in the initial HTML makes the page legible to both.

## Solution

Server-render or pre-render your primary content so it is present in the initial HTML:

- Use SSR or static generation (SSG) for the main content of each page.
- Keep purely interactive enhancements client-side — only the core content an agent needs to understand the page has to be in the raw HTML.
- Verify by fetching the page without JavaScript (e.g. `curl` the URL) and confirming the main content is present.

```bash
# What a no-JS agent sees — the main content should already be here
curl -s https://example.com/page | grep -i "your main heading"
```