> ## Documentation Index
> Fetch the complete documentation index at: https://docs.perquery.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# LLM Inspect

> See how one AI bot reads one page of your site: the screenshot it would render and the response body it receives.

The **LLM Inspect** page shows how one AI bot sees one page of your site. It shows the screenshot that the bot would render and the raw response that the bot receives. Use it to find content that a bot cannot read.

To open the page, click **LLM Inspect** in the **Analysis** group of the sidebar.

## Inspect a page

**1. Select a bot**

In the **Bot** list, click a bot. Each bot shows its platform and if it runs JavaScript (**runs JavaScript** or **no JavaScript**).

**2. Select a page**

In the **Page** list, click a page of your site. Or enter a full URL in **Any URL on this site**.

**3. Start the inspection**

Click **Inspect page**. The inspection takes a few seconds.

The **Page** list shows the pages that Perquery found on your site. If the list is empty, enter the URL by hand. You can inspect only pages on your own site.

A red dot next to a bot means that your robots.txt blocks the bot. A green dot means that the bot has access.

If a recent inspection of the same URL and bot exists, Perquery shows it and does not start a new one. The page then shows **Showing a recent run of this URL and bot; nothing was spent.**

The number of inspections depends on your plan. See [pricing](https://perquery.ai/en/pricing).

## Read the result

### Verdict

A line above the panes compares the text that the bot gets without JavaScript with the text of the fully rendered page:

| Verdict | Meaning |
| - | - |
| **server-rendered** | The bot gets almost all the text of the page in the HTML. |
| **part of the page arrives with JavaScript** | The bot does not get a part of the text without JavaScript. |
| **client-rendered only** | The bot gets very little text without JavaScript. |

Each verdict shows the number of characters that the bot sees and the number of characters on the rendered page.

> [!WARNING]
>
> A bot that does not run JavaScript cannot read the text that JavaScript adds. Put important content in the HTML that the server sends.

If your robots.txt blocks the bot, a red line says so. The panes then show what the bot would see after you unblock it. Click **you unblock it** to open the robots.txt tool. See [robots.txt](https://docs.perquery.ai/en/tools/robots-txt.md).

### Screenshot

The **Screenshot** pane shows what the bot would render.

### Raw HTML / text

The **Raw HTML / text** pane shows the response that the bot receives. The line above it shows the HTTP status, the number of redirects and the number of characters. A very long response is cut off at the stored limit.

### Failed inspections

An inspection can fail. The page then shows the cause:

* **The page could not be reached.**
* **The page did not finish loading in time.**
* **That URL is outside this site.**

Check the URL and click **Inspect page** again.

## Related pages

* [Technical analysis](https://docs.perquery.ai/en/analyze/technical-analysis.md)
* [Bot logs](https://docs.perquery.ai/en/analyze/bot-logs.md)
* [Opportunities](https://docs.perquery.ai/en/act/opportunities.md)


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.