odanielauto.com — AI Site Grade

Analysis

The Akamai Wall That Only Lets Anthropic Through

The site's Akamai WAF blocks every major AI crawler and even standard browsers with a 403 "Access Denied" -- except anthropic-ai, which receives a full 200 response with 436KB of content. This creates a bizarre situation where Claude can read the site but GPTBot, Google-Extended, PerplexityBot, and every other crawler are locked out at the edge, before even reaching the nginx origin server.

Crawler Access

Every bot except anthropic-ai is blocked at the Akamai layer. The compare_bot_access test on the homepage returned 403 for GPTBot, Google-Extended, OAI-SearchBot, ChatGPT-User, PerplexityBot, ClaudeBot, Bytespider, Applebot-Extended, and even a standard browser. Only anthropic-ai got through with status 200 and full HTML. The robots.txt itself is also behind the Akamai wall -- it returns 403 to browsers and most bots, but is accessible to anthropic-ai. Inside that robots.txt, GPTBot, OAI-SearchBot, ChatGPT-User, Claude-User, Claude-SearchBot, and PerplexityBot are all allowed to crawl the site (only blocked from JS/CSS/JSON/api/pixall paths), but the Akamai WAF overrides those permissions by blocking them entirely. No llms.txt file exists (returns 403 to browsers, 404 to anthropic-ai).

Cold-Knowledge Gap

The LLM's prior knowledge describes a completely different business: a family-owned dealership network in Omaha, Nebraska selling Honda, Hyundai, Kia, Nissan, and Mitsubishi with a "No Haggle" pricing model. The actual site is O'Daniel Automotive Group in Fort Wayne, Indiana, selling Dodge, Jeep, Audi, Mazda, Porsche, Ford, Chrysler, and Ram. The two entities share a name but have zero overlap in geography, brands, or business model. The Fort Wayne dealership has no detectable external footprint -- zero search results, zero indexed pages, zero reviews or press mentions found across multiple queries.

Schema Posture

The homepage and inventory pages carry meta name="robots" content="index, follow" tags, but those directives are meaningless when the Akamai WAF prevents Googlebot from ever reaching the page to read them. The site runs on the Dealer.com DDC platform (visible in providerID meta tags and class names like ddc-site). No JSON-LD structured data was detected in the page snippets examined -- the site relies entirely on meta tags and HTML for semantic signals. The sitemap (accessible only to anthropic-ai) contains 166KB of URLs with lastmod dates in May 2026, suggesting a future-dated publishing system or a platform bug.

External Signals

The site has zero detectable external presence. Searches for the business name, domain, and location returned no results across web search, reviews, or social mentions. The domain uses Microsoft 365 for email (Exchange Online MX records) and is hosted on Akamai with an origin nginx server. The complete absence of indexed pages means AI models have no secondary sources to triangulate against -- the only way to learn about this dealership is through the anthropic-ai backdoor, which no other major AI crawler can exploit.

Findings

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

   The Akamai Web Application Firewall returns 403 Access Denied for GPTBot, Google-Extended, OAI-SearchBot, ChatGPT-User, PerplexityBot, ClaudeBot, Bytespider, Applebot-Extended, and standard browsers. Only anthropic-ai receives a 200 response with full HTML content.

   What to change: Reconfigure the Akamai WAF to allow access for all major AI crawlers (GPTBot, Google-Extended, OAI-SearchBot, ChatGPT-User, PerplexityBot, ClaudeBot, etc.) to the homepage and key content paths.

2. Robots.txt is blocked by Akamai for most bots High

   The robots.txt file returns 403 to browsers and most bots, but is accessible to anthropic-ai. Inside, GPTBot, OAI-SearchBot, ChatGPT-User, Claude-User, Claude-SearchBot, and PerplexityBot are allowed to crawl the site (only blocked from JS/CSS/JSON/api/pixall paths), but the Akamai WAF overrides those permissions by blocking them entirely.

   What to change: Ensure robots.txt is publicly accessible without authentication, and remove Akamai WAF rules that block bots from reading it.

3. No llms.txt file exists Medium

   The llms.txt file returns 403 to browsers and 404 to anthropic-ai, meaning no AI-friendly summary of the site is available.

   What to change: Create an llms.txt file at the root that provides a concise summary of the dealership, its locations, brands, and key pages for AI crawlers.

4. LLM prior knowledge describes a different business High

   The LLM's prior knowledge describes O'Daniel Automotive Group as a family-owned dealership network in Omaha, Nebraska selling Honda, Hyundai, Kia, Nissan, and Mitsubishi with a 'No Haggle' pricing model. The actual site is O'Daniel Automotive Group in Fort Wayne, Indiana, selling Dodge, Jeep, Audi, Mazda, Porsche, Ford, Chrysler, and Ram. The two entities share a name but have zero overlap in geography, brands, or business model.

   What to change: Publish accurate business information across the web (Google Business Profile, dealer directories, press releases) to correct the LLM's knowledge and establish the correct identity.

5. Zero external presence across web search and reviews High

   Multiple web searches for the business name, domain, and location returned zero results. No indexed pages, no reviews, no social mentions, and no press coverage were found. The domain has no detectable external footprint.

   What to change: Build an external presence by creating and verifying a Google Business Profile, listing on dealer directories, and encouraging customer reviews on platforms like Google, DealerRater, and Cars.com.

6. No JSON-LD structured data detected Medium

   The site relies entirely on meta tags and HTML for semantic signals; no JSON-LD structured data was detected in the page snippets examined. This limits the ability of AI crawlers to extract entity relationships and attributes.

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

7. Sitemap contains future-dated lastmod timestamps Medium

   The sitemap (accessible only to anthropic-ai) contains lastmod dates in May 2026, suggesting a future-dated publishing system or a platform bug. This may confuse crawlers about content freshness.

   What to change: Correct the sitemap lastmod dates to reflect actual publication dates, and ensure the sitemap is publicly accessible.

8. Sitemap is blocked by Akamai for most bots High

   The sitemap.xml returns 403 to standard browsers and most bots, but is accessible to anthropic-ai. This prevents search engines from discovering the site's URL structure.

   What to change: Make the sitemap publicly accessible by removing Akamai WAF rules that block it.

9. Homepage returns 403 to standard browsers High

   The homepage at https://www.odanielauto.com returns 403 Access Denied to standard browsers, making the site inaccessible to human visitors and most crawlers.

   What to change: Reconfigure the Akamai WAF to allow standard browser access to the homepage and all public content.

10. About page returns 404 Medium

    The about page at /about/index.htm returns a 404 status code when fetched as anthropic-ai, indicating a broken link or missing content.

    What to change: Fix the about page URL or create a proper about page with business information.

What's working

• Anthropic AI crawler is allowed through Akamai — The anthropic-ai crawler receives a 200 response with full HTML content, allowing Claude to access the site's content despite the WAF blocking other bots.
• Robots.txt allows major AI crawlers — The robots.txt file (when accessible) permits GPTBot, OAI-SearchBot, ChatGPT-User, Claude-User, Claude-SearchBot, and PerplexityBot to crawl the site, only blocking JS/CSS/JSON/api/pixall paths.
• Meta robots tags set to index, follow — The homepage and inventory pages carry meta name='robots' content='index, follow' tags, indicating intent for pages to be indexed.
• Sitemap exists and is accessible to anthropic-ai — A sitemap.xml file exists with 166KB of URLs, accessible to the anthropic-ai crawler, providing a comprehensive list of site pages.
• Dealer.com DDC platform provides structured markup — The site runs on the Dealer.com DDC platform, which typically includes structured data and meta tags for inventory, though JSON-LD was not detected in the snippets examined.