AI Site Grade

carbonhealth.com — AI Site Grade

Carbon Health's entire domain returns 403 to every AI crawler, making the site a black box to the AI ecosystem.

Carbon Health's live site is completely blocked from all AI crawlers, with no robots.txt, sitemap, or llms.txt accessible, and zero search engine visibility.

Findings
11
Evidence checks
35
Completed
30 May 2026

Analysis

The live site is entirely invisible to every AI crawler tested

Every AI crawler — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, ChatGPT-User, OAI-SearchBot, Applebot-Extended, and anthropic-ai — receives a 403 Cloudflare block on the homepage, with zero content delivered. The robots.txt and llms.txt endpoints also return 403. The sitemap is unreachable. No AI crawler can access any page on carbonhealth.com. The site is a black box to the entire AI ecosystem.

Crawler Access

The site sits behind Cloudflare with no bypass for any known AI user-agent. The compare_bot_access test confirmed all 11 bots tested receive identical 403 responses with Cloudflare's "Attention Required" block page. No robots.txt rules exist because the file itself is blocked. No llms.txt exists. The DNS records show Cloudflare nameservers (brit.ns.cloudflare.com, skip.ns.cloudflare.com) and the server header confirms Cloudflare. The Wayback Machine snapshot from December 2024 shows the site was previously accessible, meaning the blanket blocking is a recent or ongoing configuration.

Cold-Knowledge Gap

The LLM model knows Carbon Health as a hybrid healthcare chain founded in 2015 by Eren Bali and Dr. Caesar Djavaherian, with over 100 clinics, $500M+ funding, and a 2023 layoff of ~200 jobs. The model also recalls clinic closures and billing complaints. The live site — which the model cannot access — presents a polished, patient-friendly brand emphasizing "care as it should be," same-day appointments, and transparent pricing. The gap between the model's stored knowledge (financial strain, layoffs) and the site's aspirational messaging is stark. The model cannot verify whether the site still lists 100+ clinics or has updated its service footprint.

Schema Posture

The archived homepage and key pages contain MedicalOrganization, Organization, and WebSite schema with a San Francisco address (300 California St. 7th Floor), phone number +1-415-612-3256, and email [email protected]. The schema is present but minimal — no FAQPage, Product, LocalBusiness, or Service schema for individual clinic locations or specific medical services. The urgent care page has an FAQ section with questions like "Do you accept my insurance?" but no corresponding FAQPage schema. The pricing page lists 42+ line-item prices in a table format but uses no PriceSpecification or MedicalProcedure schema.

External Signals

The search tool returned zero results for Carbon Health across multiple queries, which is anomalous. The LLM's cold knowledge references layoffs and clinic closures from 2023-2024, but no recent press or Reddit threads surfaced. The Wayback snapshot shows the site was live and accessible as recently as December 2024 with a full service page structure. The blog footer copyright reads "©2022," suggesting the content may not have been updated recently. The site's complete invisibility to search and AI crawlers means external reputation signals are entirely disconnected from the live domain.

Findings

  1. All AI crawlers receive 403 Cloudflare block on homepage High

    Every AI crawler tested (GPTBot, ClaudeBot, PerplexityBot, Google-Extended, ChatGPT-User, OAI-SearchBot, Applebot-Extended, anthropic-ai) receives a 403 Cloudflare block on the homepage with zero content delivered. The site is entirely invisible to the AI ecosystem.

    What to change: Configure Cloudflare to allow known AI crawler user-agents by creating a firewall rule that permits GPTBot, ClaudeBot, PerplexityBot, Google-Extended, ChatGPT-User, OAI-SearchBot, Applebot-Extended, and anthropic-ai to access the site.

  2. robots.txt returns 403, no crawler directives available High

    The robots.txt endpoint returns a 403 Cloudflare block, preventing any crawler from reading crawl directives. No user-agent rules exist because the file itself is inaccessible.

    What to change: Unblock robots.txt from Cloudflare so crawlers can read crawl directives. Add rules to allow AI crawlers and disallow any private paths.

  3. Sitemap returns 403, no URL discovery possible High

    The sitemap endpoint returns a 403 Cloudflare block, preventing crawlers from discovering site URLs. The list_known_urls tool found zero URLs.

    What to change: Unblock the sitemap and ensure it lists all important pages. Submit the sitemap to Google Search Console and Bing Webmaster Tools.

  4. llms.txt endpoint returns 403, no AI guidance file Medium

    The llms.txt endpoint returns a 403 Cloudflare block, meaning no guidance file exists for AI crawlers to understand site structure or content priorities.

    What to change: Create an llms.txt file that lists key pages and provides a brief description of the site for AI crawlers.

  5. Zero search results for carbonhealth.com across multiple queries High

    Multiple web searches for carbonhealth.com and related terms returned zero results, indicating the site is not indexed by search engines. This is anomalous for an established healthcare brand.

    What to change: Investigate why the site is not indexed. Ensure Cloudflare is not blocking Googlebot and Bingbot. Submit the site to Google Search Console and Bing Webmaster Tools.

  6. FAQ content on urgent care page lacks FAQPage schema Medium

    The urgent care page contains an FAQ section with questions like 'Do you accept my insurance?' but no FAQPage structured data is present. This misses an opportunity for rich results in AI and search.

    What to change: Add FAQPage schema markup to the FAQ section on the urgent care page.

  7. Pricing page uses plain table without PriceSpecification schema Medium

    The insurance and pricing page lists 42+ line-item prices in a table format but uses no PriceSpecification or MedicalProcedure schema, reducing machine readability.

    What to change: Add PriceSpecification and MedicalProcedure schema to each pricing line item.

  8. Key pages have minimal schema, missing LocalBusiness and Service types Medium

    Archived pages contain MedicalOrganization, Organization, and WebSite schema but lack LocalBusiness, Service, or Product schema for individual clinic locations or specific medical services.

    What to change: Add LocalBusiness schema for each clinic location and Service schema for each medical service offered.

  9. Blog footer copyright reads 2022, suggesting outdated content Low

    The blog page footer displays '©2022', indicating the blog content may not have been updated recently. This can affect trust and freshness signals.

    What to change: Update the blog footer copyright to the current year and ensure blog content is regularly refreshed.

  10. LLM cold knowledge contains negative signals not reflected on live site Medium

    The LLM model recalls layoffs, clinic closures, and billing complaints for Carbon Health, while the live site presents a polished brand. The model cannot verify current clinic count or service footprint, creating a disconnect.

    What to change: Proactively publish positive news, updated clinic locations, and patient testimonials to counterbalance negative signals. Ensure the site is accessible to AI crawlers so they can index current content.

  11. No recent press or Reddit discussions found for Carbon Health Low

    Web searches for Carbon Health reviews, layoffs, and news returned zero results, indicating a lack of recent external signals. This may be due to the site's invisibility to search engines.

    What to change: Encourage patient reviews on third-party platforms and engage in PR to generate positive press coverage.

What's working

  • Homepage and key pages contain MedicalOrganization and Organization schema — The archived homepage and about page include MedicalOrganization, Organization, and WebSite schema with accurate contact information (address, phone, email), providing a foundation for entity recognition.
  • Wayback Machine snapshot from December 2024 shows previously accessible site — A Wayback Machine snapshot from December 2024 confirms the site was previously accessible with a full service page structure, indicating the current blocking is a recent configuration change.
  • Pricing page lists 42+ line-item prices in a clear table — The insurance and pricing page provides transparent pricing for over 42 medical services in a table format, which is valuable for patients and could be enhanced with schema.
  • Urgent care page includes helpful FAQ section — The urgent care page contains an FAQ section addressing common patient questions, which is useful content that could be enhanced with FAQPage schema.
  • Blog page exists with 379 words of content — The blog page contains content about Carbon Health services, providing a channel for fresh content and SEO if made accessible.

Track carbonhealth.com across AI search

This is one snapshot. Open the interactive report to inspect evidence, or grade another site free.

Open this AI Site Grade Grade another site Track your brand