maxmotors.com — AI Site Grade

Analysis

Anthropic-Only Access: The Akamai Gate

Max Motors Auto Group operates a site that returns 403 Access Denied to every browser and every major AI crawler except anthropic-ai, which receives a full 200 with 523KB of rendered HTML. This creates a bizarre inversion: Claude can read the entire inventory, but Google-Extended, GPTBot, PerplexityBot, and human visitors all hit an Akamai firewall wall.

Crawler Access

The robots.txt at www.maxmotors.com/robots.txt (accessible only to anthropic-ai) explicitly disallows GPTBot, OAI-SearchBot, ChatGPT-User, Claude-User, Claude-SearchBot, and PerplexityBot from JS, CSS, JSON, and API paths — but does not disallow them from the homepage or inventory pages. The Akamai WAF blocks them anyway at the network layer. The site runs on dealer.com (DDC platform) behind AkamaiGHost, with a secondary nginx origin. The anthropic-ai user-agent bypasses Akamai entirely and hits nginx directly, while all other traffic is terminated at the edge.

llms.txt Posture

The site has a 145KB llms.txt — unusually large for a dealership — containing hundreds of markdown link entries for individual vehicle inventory pages. The file is served with max-age=86400 cache control and is accessible only to anthropic-ai. Every entry uses a generic description ("Browse our inventory of Dodge, Jeep, Chevrolet, Ford, Chrysler, GMC, Ram vehicles for sale at Max Motors Auto Group") rather than vehicle-specific summaries, making it a low-signal directory dump rather than a curated AI resource.

Cold-Knowledge Gap

LLM knowledge describes Max Motors as a Missouri dealership group in Butler and Harrisonville with "No Haggle" pricing, selling Ford, Chevrolet, Chrysler, Dodge, Jeep, and Ram. The site itself reveals Hyundai as an additional brand (mentioned in the new-inventory page title) and identifies the entity as "Max Motors Auto Group" — a name the cold knowledge does not use. The site's actual content emphasizes Harrisonville as the primary location, while cold knowledge lists Butler first. The "No Haggle" policy that LLMs consider a defining feature is not prominently signaled in the site's page titles or meta descriptions.

Content & Schema

The homepage and inventory pages are heavy JavaScript applications (DDC platform with Webpack bundles for facets, filters, listing data). The homepage title reads "Max Motors | New & Used Cars & Trucks in the Midwest" with a meta description mentioning "Chrysler, Chevrolet, Ford, Dodge, and GMC vehicles in the Harrisonville area." The about-us.htm page returns a 404 — the dealership has no dedicated about page. The hyundai-inventory path also 404s, suggesting Hyundai may be a discontinued brand line or an unlinked page. No JSON-LD schema was detected on any fetched page. The site uses index, follow robots meta tags on all pages, but the Akamai block prevents Google from ever reading them.

Findings

1. Akamai WAF blocks all AI crawlers except anthropic-ai High

   The site returns 403 Access Denied to Google-Extended, GPTBot, PerplexityBot, and all other major AI crawlers. Only anthropic-ai receives a 200 response, creating a single-vendor AI visibility channel.

   What to change: Configure the Akamai WAF to allow Google-Extended, GPTBot, PerplexityBot, and other major AI crawlers to access the site, or implement a bot management solution that selectively permits AI crawlers.

2. robots.txt disallows GPTBot and others from JS, CSS, and API paths Medium

   The robots.txt file explicitly disallows GPTBot, OAI-SearchBot, ChatGPT-User, Claude-User, Claude-SearchBot, and PerplexityBot from accessing JS, CSS, JSON, and API paths. While the Akamai block makes this moot, it signals intent to restrict AI crawlers.

   What to change: Remove the disallow directives for AI crawlers from robots.txt, or update them to allow access to all paths if the WAF is configured to permit those bots.

3. 145KB llms.txt contains generic inventory dumps with no vehicle-specific summaries Medium

   The llms.txt file is 145KB with hundreds of markdown link entries for individual vehicle pages, but every entry uses the same generic description. This provides low signal to AI models and is not a curated resource.

   What to change: Replace the generic descriptions with vehicle-specific summaries (make, model, year, key features, price) and reduce the file size by linking only to top-level inventory categories rather than individual VDPs.

4. llms.txt is only accessible to anthropic-ai due to Akamai block High

   The llms.txt file is served only to anthropic-ai; all other AI crawlers receive a 403. This defeats the purpose of llms.txt as a cross-model resource.

   What to change: Ensure the llms.txt file is accessible to all AI crawlers by updating the Akamai WAF rules to allow access to /llms.txt for all user-agents.

5. About Us page returns 404 error Medium

   The /about-us.htm page returns a 404 status code, meaning the dealership has no dedicated about page. This deprives AI crawlers and users of background information about the business.

   What to change: Create an about page with details about the dealership's history, locations, and values, and ensure it is accessible to all crawlers.

6. Hyundai inventory path returns 404, suggesting discontinued brand line Low

   The /hyundai-inventory/hyundai-vehicles-for-sale.htm path returns a 404, indicating that Hyundai may no longer be sold or the page is unlinked. This creates a broken link experience for AI crawlers that may have indexed the path.

   What to change: Either restore the Hyundai inventory page or remove the path and ensure no internal links point to it.

7. No JSON-LD structured data detected on any page High

   No JSON-LD schema was found on the homepage, inventory pages, or contact page. This means AI crawlers cannot easily extract structured information about the dealership, inventory, or location.

   What to change: Add JSON-LD structured data for AutoDealer, VehicleInventory, and LocalBusiness schemas to all relevant pages.

8. Cold knowledge lists Butler as primary location and misses Hyundai brand Medium

   LLM knowledge describes Max Motors as based in Butler, Missouri, but the site emphasizes Harrisonville. The site also sells Hyundai, which is not mentioned in cold knowledge. This mismatch can confuse AI models.

   What to change: Ensure the site prominently and consistently states the primary location and all brands sold in page titles, meta descriptions, and structured data.

9. No Haggle pricing policy not prominently signaled in page titles or meta descriptions Low

   The 'No Haggle' policy that LLMs associate with Max Motors is not featured in page titles or meta descriptions, reducing the chance that AI models will surface this differentiator.

   What to change: Include 'No Haggle Pricing' in the homepage title and meta description, and consider adding it to structured data.

10. Homepage and inventory pages are heavy JavaScript applications Medium

    The site relies on JavaScript (Webpack bundles) to render content, which can hinder AI crawlers that do not execute JavaScript. Although anthropic-ai receives rendered HTML, other bots may see empty shells.

    What to change: Implement server-side rendering (SSR) or dynamic rendering to serve static HTML to crawlers, ensuring content is accessible without JavaScript.

11. Sitemap returns 403 to most crawlers High

    The sitemap at /sitemap.xml is only accessible to anthropic-ai; other crawlers receive a 403. This prevents search engines and AI crawlers from discovering the site's URL structure.

    What to change: Configure the Akamai WAF to allow all crawlers to access /sitemap.xml.

12. robots.txt returns 403 to most crawlers High

    The robots.txt file is only accessible to anthropic-ai; other crawlers receive a 403. This violates the robots exclusion protocol and prevents crawlers from understanding crawl rules.

    What to change: Configure the Akamai WAF to allow all crawlers to access /robots.txt.

What's working

• Anthropic's Claude receives full rendered HTML with 523KB of content — The anthropic-ai bot bypasses Akamai and receives a 200 response with fully rendered HTML, including inventory listings and page content. This ensures Claude can access the site's information.
• Site provides a 145KB llms.txt file with hundreds of inventory links — The llms.txt file contains hundreds of markdown links to individual vehicle pages, providing a broad directory of inventory for AI models that can access it.
• All pages use index, follow robots meta tags — The site uses 'index, follow' meta tags on all pages, signaling that content should be indexed and links followed. This is a positive signal for crawlers that can access the pages.
• Contact page is accessible and returns 200 for anthropic-ai — The /contact.htm page returns a 200 status with 270KB of content, providing contact information for the dealership.
• New and used inventory pages return 200 for anthropic-ai — The new-inventory and used-inventory index pages return 200 status with substantial HTML content (390KB and 383KB respectively), indicating the inventory is accessible to Claude.
• Financing and service pages return 200 for anthropic-ai — The financing and service index pages return 200 status with content (311KB and 278KB), providing information about financing options and service departments.
• Site runs on dealer.com (DDC) platform with standard infrastructure — The site uses the dealer.com platform, which is a common and well-supported automotive dealership CMS. This provides a solid foundation for potential improvements.
• Homepage has a descriptive title and meta description — The homepage title 'Max Motors | New & Used Cars & Trucks in the Midwest' and meta description mentioning brands and location provide basic SEO signals.