jimbutlerautogroup.com — AI Site Grade

Analysis

Cloudflare Blocks All Browsers and Major AI Crawlers While Selectively Admitting ChatGPT-User and Perplexity-User

The site's Cloudflare WAF returns HTTP 403 to browsers, GPTBot, ClaudeBot, Google-Extended, PerplexityBot, Applebot-Extended, and Bytespider, but delivers a full 244 KB HTML page to ChatGPT-User, OAI-SearchBot, and Perplexity-User. This creates a bizarre inversion: the most widely used AI training crawlers are locked out, while the search/chat bots from the same companies are let in.

Crawler Access

The robots.txt (accessible only via bot user-agents) contains no rules for GPTBot, ClaudeBot, Google-Extended, or PerplexityBot. It disallows /gallery/, /eprice/, /print-auto/, and various search-parameter patterns, but allows /new/?s:pr=1&ct=20&tp=new and /search/used/?s:pr=1&ct=20&tp=used. The llms.txt returns a Cloudflare 403 block — the file does not exist. The sitemap index at /resrc/xmlsitemap/xml-sitemaps/ points to two sub-sitemaps containing thousands of inventory listing URLs, all of which are also Cloudflare-gated for standard crawlers.

Cold-Knowledge Gap

The LLM knows Jim Butler Auto Group as a family-owned chain in Linn, Cuba, and St. James, Missouri selling Chevrolet, Buick, GMC, and CDJR brands. The actual site reveals a much larger operation: 11 locations spanning Fenton, Chesterfield, Maplewood, St. Louis, Linn, and Centralia, selling nine brands including Alfa Romeo, Maserati, Kia, and FIAT. The cold knowledge is missing the Fenton headquarters entirely, omits the luxury brands, and describes a smaller rural footprint than the site claims. The gap is substantial — the LLM describes a different dealership group than what the site presents.

Schema Posture

The homepage contains five AutoDealer JSON-LD blocks covering Jim Butler Chevrolet Fenton, Jim Butler Maserati, Jim Butler Alfa Romeo, Jim Butler Chevrolet Linn, and Jim Butler CDJR Linn. Each includes PostalAddress, GeoCoordinates, and a brand array listing all nine makes. However, the schema omits several known locations: Jim Butler Kia (Chesterfield), Jim Butler FIAT (St. Louis), Jim Butler Chevrolet Centralia, Jim Butler Outlet, and Jim Butler Centralia CDJR. The name field for the Maserati location contains a trailing comma in the address (Maplewood,). No FAQPage, Product, or Service schema was detected on any fetched page.

External Signals

DuckDuckGo returns zero search results for the brand name, the domain, or any location-specific query. No Google reviews, Reddit threads, press mentions, or third-party citations were found through available search tools. The site itself hosts a blog with over 1.4 MB of HTML, but no external backlink ecosystem is visible. This absence of off-domain signals means AI models have almost no third-party corroboration to draw on when describing the dealership.

Technical Fragility

The site runs on the DealerEprocess platform behind Cloudflare, serving XHTML 1.0 Transitional doctype. Page sizes are enormous (200-800+ KB) due to inline JavaScript and CSS bundles. The find-my-car page alone is 848 KB. The robots.txt Crawl-delay: 7 is unusually aggressive. The sitemap lastmod dates show 2026-05-30, suggesting either a future-dated cache issue or a platform quirk. No canonical issues were found, but the site's dependency on JS-rendered navigation means even permitted bots may miss content if they do not execute JavaScript.

Findings

1. Cloudflare WAF blocks all browsers and major AI crawlers High

   The site returns HTTP 403 to browsers, GPTBot, ClaudeBot, Google-Extended, PerplexityBot, Applebot-Extended, and Bytespider, while delivering full HTML to ChatGPT-User, OAI-SearchBot, and Perplexity-User. This blocks the most widely used AI training crawlers.

   What to change: Adjust Cloudflare WAF rules to allow standard browsers and major AI crawlers (GPTBot, ClaudeBot, Google-Extended, PerplexityBot, Applebot-Extended, Bytespider) to access the site, while maintaining security for other traffic.

2. Robots.txt contains no rules for major AI crawlers Medium

   The robots.txt file (accessible only via bot user-agents) has no directives for GPTBot, ClaudeBot, Google-Extended, or PerplexityBot, leaving their access unmanaged.

   What to change: Add explicit allow/disallow rules for GPTBot, ClaudeBot, Google-Extended, and PerplexityBot in robots.txt.

3. llms.txt file returns 403 and does not exist Medium

   The llms.txt file is not published; requests return a Cloudflare 403 block.

   What to change: Create and publish an llms.txt file with a summary of the site's content and links to key pages for AI assistants.

4. LLM cold knowledge omits headquarters, luxury brands, and multiple locations High

   The LLM knows Jim Butler Auto Group as a small rural chain selling Chevrolet, Buick, GMC, and CDJR, but the site shows 11 locations, 9 brands including Alfa Romeo and Maserati, and a Fenton headquarters. The cold knowledge describes a different, smaller dealership group.

   What to change: Improve off-site signals (press releases, directory listings, social media) and on-site schema to ensure AI models have accurate, comprehensive information about the dealership group.

5. AutoDealer schema omits five known locations High

   The homepage includes five AutoDealer JSON-LD blocks but omits Jim Butler Kia (Chesterfield), Jim Butler FIAT (St. Louis), Jim Butler Chevrolet Centralia, Jim Butler Outlet, and Jim Butler Centralia CDJR.

   What to change: Add AutoDealer schema blocks for all missing locations, ensuring each includes accurate name, address, phone, and geo coordinates.

6. Maserati location schema has trailing comma in address Low

   The AutoDealer schema for Jim Butler Maserati contains a trailing comma in the address field ("Maplewood,"), which may cause parsing issues.

   What to change: Remove the trailing comma from the address field in the Maserati AutoDealer schema.

7. No FAQPage, Product, or Service schema detected on any page Medium

   Despite having blog posts and service pages, no FAQPage, Product, or Service structured data was found, limiting rich result eligibility.

   What to change: Add FAQPage schema to blog posts and Product/Service schema to relevant pages to enhance AI understanding and rich snippet potential.

8. Zero search results for brand name or domain on DuckDuckGo High

   Multiple searches for the brand name, domain, and location-specific queries returned zero results, indicating no external backlink ecosystem or third-party citations.

   What to change: Build external signals through press releases, local business directories, social media profiles, and backlink acquisition to improve AI model corroboration.

9. Page sizes are excessively large (200-848 KB) due to inline assets Medium

   Pages contain large inline JavaScript and CSS bundles, with the find-my-car page at 848 KB and blog at 1.4 MB, slowing load times and potentially affecting crawl budget.

   What to change: Minify and defer JavaScript, inline critical CSS only, and lazy-load non-essential resources to reduce page weight.

10. Robots.txt sets aggressive Crawl-delay of 7 seconds Low

    The Crawl-delay: 7 directive may slow down search engine crawling and reduce indexation frequency.

    What to change: Reduce Crawl-delay to a lower value (e.g., 1-2 seconds) or remove it to allow faster crawling.

11. Sitemap lastmod dates show future dates (2026-05-30) Medium

    The sitemap index and sub-sitemaps list lastmod dates of 2026-05-30, which may confuse crawlers and indicate a platform caching issue.

    What to change: Correct the sitemap lastmod dates to reflect actual modification times and investigate the platform's date handling.

12. Site relies on JavaScript-rendered navigation, potentially hiding content from bots Medium

    Even permitted bots that do not execute JavaScript may miss navigation links and content, limiting their ability to discover pages.

    What to change: Ensure critical navigation and content are available in the initial HTML response, using server-side rendering or progressive enhancement.

What's working

• ChatGPT-User, OAI-SearchBot, and Perplexity-User receive full HTML content — The site correctly serves full 244 KB HTML pages to ChatGPT-User, OAI-SearchBot, and Perplexity-User, enabling AI search and chat features to read the content.
• Homepage includes five AutoDealer JSON-LD blocks with brand arrays — The homepage contains structured data for five dealership locations, each with address, geo coordinates, and a brand array listing all nine makes, providing rich context for AI models.
• Sitemap index and inventory sub-sitemaps are accessible to permitted bots — The sitemap index at /resrc/xmlsitemap/xml-sitemaps/ and sub-sitemaps containing thousands of inventory URLs are served to OAI-SearchBot, aiding discovery.
• Blog page contains over 1.4 MB of HTML content — The blog page at /blog/ has substantial text content (1.4 MB HTML), providing a rich source of information for AI models that can access it.
• Robots.txt allows new and used inventory search pages — The robots.txt explicitly allows paths like /new/?s:pr=1&ct=20&tp=new and /search/used/?s:pr=1&ct=20&tp=used, enabling crawling of inventory listings.
• No canonical tag issues found on fetched pages — The pages checked did not exhibit canonical tag problems, reducing the risk of duplicate content confusion.
• Site has a recent Wayback Machine snapshot (2026-03-08) — A Wayback Machine snapshot from March 2026 captures the homepage, providing a fallback for historical content.