---
title: "agent-ready.dev adds an AI snippet check | Agent Readiness Compare"
description: "2 September 2026: agent-ready.dev added check P25, which looks for max-snippet:-1 and max-image-preview:large in meta robots."
canonical: https://agentreadinesscompare.com/news/2026-09-02-agent-ready-dev-adds-a-check-for-ai/
site: "Agent Readiness Compare"
last_updated: 2026-11-05
date: 2026-09-02
section: "News"
---

# agent-ready.dev adds an AI snippet check | Agent Readiness Compare

> 2 September 2026: agent-ready.dev added check P25, which looks for max-snippet:-1 and max-image-preview:large in meta robots.

Source: https://agentreadinesscompare.com/news/2026-09-02-agent-ready-dev-adds-a-check-for-ai/
Site: Agent Readiness Compare
Last reviewed: 2026-11-05

# agent-ready.dev adds a check for AI snippet directives

In short

agent-ready.dev added check P25, which looks for max-snippet:-1 and max-image-preview:large in meta robots or X-Robots-Tag, the directives that govern how much of a page Google may show in AI Overviews. agent-ready.dev's methodology says its overall score is passing checks divided by total checks, so warnings and failures both count against a site.

## 1. What happened

agent-ready.dev added check P25, which looks for max-snippet:-1 and max-image-preview:large in meta robots or X-Robots-Tag, the directives that govern how much of a page Google may show in AI Overviews. [1](#src-1)

Date: 2 September 2026. Source type: a primary source (the company or standard body itself).

## 2. Who it involves

-   [agent-ready.dev](https://agentreadinesscompare.com/tools/agent-ready-dev/)Scanner that grades a site on 72 checks, 69 mapped to published specs (Vercel's agent readability guide, llmstxt.org and agent protocol specs), with an llms.txt sub-score and a separate MCP server grader.

## 3. Nearby items

-   [28 August 2026: Vercel publishes its guide to making sites readable by agents](https://agentreadinesscompare.com/news/2026-08-28-vercel-publishes-its-guide-to-making-sites-readable/)
-   [20 October 2026: agent-ready.dev stops crediting catch-all pages as llms.txt](https://agentreadinesscompare.com/news/2026-10-20-agent-ready-dev-stops-crediting-catch-all-pages/)
-   [4 November 2026: agent-ready.dev updates its comparison with platform scanners](https://agentreadinesscompare.com/news/2026-11-04-agent-ready-dev-updates-its-comparison-with-platform/)

## 4. More from the same source

-   agent-ready.dev's methodology says its overall score is passing checks divided by total checks, so warnings and failures both count against a site.[1](#src-1)

## 5. Related

-   [News index](https://agentreadinesscompare.com/news/)

## 6. Sources

1.  agent-ready.dev methodology. [https://agent-ready.dev/methodology](https://agent-ready.dev/methodology), read 2026-11-05.

Last reviewed 2026-11-05
