tricolor.com — AI Site Grade

Analysis

Bankruptcy-Notice Portal Behind a Login Wall, AI Crawlers Blocked

The public-facing tricolor.com no longer serves any marketing or brand content — it is a bankruptcy customer-service portal behind a redirect to my.tricolor.com/?page=help, and every major AI crawler is explicitly blocked from accessing it.

Crawler Access

The robots.txt uses Cloudflare's managed content-signal framework, setting Content-Signal: search=yes,ai-train=no for User-agent: *, then separately disallowing GPTBot, ClaudeBot, Google-Extended, Applebot-Extended, Bytespider, CCBot, and meta-externalagent from the entire site. Actual bot-access testing confirms: GPTBot, ClaudeBot, ChatGPT-User, OAI-SearchBot, PerplexityBot, Perplexity-User, Bytespider, and anthropic-ai all receive HTTP 403 from Cloudflare. Only Google-Extended and Applebot-Extended return 200 — but they land on the same login-wall page as a browser, with no structured data, no canonical URL, and a generic <meta name="description" content="Description goes here.">. The llms.txt URL resolves to the same login page (no actual llms.txt file exists). The sitemap is empty (0 URLs).

Cold-Knowledge Gap

LLM knowledge about Tricolor Holdings describes a used-car retailer serving Hispanic consumers with in-house financing and a credit-builder program, headquartered in Irving, Texas, and owned by Grupo Financiero Banorte. It mentions a 2023 CFPB lawsuit. The live site contains none of this. The site's only content is a bilingual FAQ explaining that Tricolor filed for bankruptcy, loan servicing has been transferred to Vervent, and customers must continue payments. The brand's pre-bankruptcy identity — dealerships, vehicle inventory, credit-building products — is entirely absent from the public web presence.

Content and Schema Posture

The homepage and FAQ page contain zero JSON-LD schema of any type. No Organization, FAQPage, Product, or WebSite schema is present. The FAQ content uses plain Q: / A: text formatting with no semantic markup. The page title is "Login" or "Customer Service Center." The meta description is a placeholder string. Heading structure uses only H3 tags, with no H1 or H2 elements. The site is hosted on Cloudflare with HSTS, X-Frame-Options: SAMEORIGIN, and X-Content-Type-Options: nosniff.

External Signals

Web search returns zero indexed results for any query involving tricolor.com, Tricolor Holdings, or Tricolor auto sales — no press coverage, no reviews, no Reddit threads, no BBB listings, no CFPB lawsuit references appear in search results. The site has no external backlinks detected. This complete search invisibility, combined with the empty sitemap and login-wall redirect, means the domain has been effectively de-indexed. The only external signal is the LLM's prior knowledge, which describes a company that no longer matches the site's current reality.

Findings

1. All major AI crawlers blocked by robots.txt and Cloudflare High

   Robots.txt disallows GPTBot, ClaudeBot, Google-Extended, Applebot-Extended, Bytespider, CCBot, and meta-externalagent. Bot access tests confirm 403 for most AI bots; only Google-Extended and Applebot-Extended return 200 but land on a login wall.

   What to change: Remove disallow rules for AI crawlers or serve a public-facing version of the site with brand content.

2. Public site redirects to login wall with no brand content High

   The homepage redirects to my.tricolor.com/?page=help, a bankruptcy customer-service portal. No marketing, inventory, or brand information is publicly accessible.

   What to change: Restore a public-facing website with brand information, vehicle inventory, and company details.

3. Zero JSON-LD structured data on any page High

   The homepage and FAQ page contain no JSON-LD schema of any type (Organization, FAQPage, Product, WebSite). FAQ content uses plain Q:/A: text without semantic markup.

   What to change: Add JSON-LD structured data for Organization, FAQPage, and WebSite on all pages.

4. Sitemap returns zero URLs High

   The sitemap at tricolor.com/sitemap.xml returns 200 but contains no URLs, making it impossible for search engines to discover pages.

   What to change: Generate and submit a valid sitemap listing all public pages.

5. llms.txt resolves to login page, not a real file Medium

   The llms.txt URL returns the same login page content instead of a proper llms.txt file, providing no guidance to AI crawlers.

   What to change: Create a valid llms.txt file with pointers to public content.

6. Meta description is a placeholder string Medium

   The homepage meta description reads 'Description goes here.', providing no useful summary for search engines or AI crawlers.

   What to change: Write a descriptive meta description summarizing the site's purpose.

7. No H1 or H2 headings on any page Medium

   The homepage and FAQ page use only H3 tags, with no H1 or H2 elements, harming content hierarchy and accessibility.

   What to change: Add a single H1 heading and use H2 for section titles.

8. Domain has zero indexed pages in search engines High

   Web searches for tricolor.com and related queries return no results. The site has no backlinks, no press coverage, and no external signals.

   What to change: Restore public content, submit sitemap, and build backlinks to regain indexing.

9. LLM knowledge describes a different company than the live site Medium

   LLMs recall Tricolor as a used-car retailer with in-house financing, but the live site only shows a bankruptcy portal. This mismatch can cause AI-generated summaries to be inaccurate.

   What to change: Publish current brand information on the public site to align with LLM knowledge.

10. FAQ content uses plain text instead of semantic markup Low

    The FAQ page formats questions and answers as plain Q:/A: text with no HTML list or schema markup, reducing its utility for AI crawlers.

    What to change: Use <dl> or <details> elements and add FAQPage schema.

What's working

• Strong security headers in place — The site uses HSTS, X-Frame-Options: SAMEORIGIN, and X-Content-Type-Options: nosniff, providing good security posture.
• Bilingual FAQ provides clear bankruptcy guidance — The FAQ page offers both English and Spanish content explaining the bankruptcy process and loan servicing transfer to Vervent, serving affected customers.
• Robots.txt uses Cloudflare content-signal framework — The robots.txt sets Content-Signal: search=yes,ai-train=no for User-agent: *, a modern approach to controlling AI training access.