creativefabrica.com — AI Site Grade

Analysis

Creative Fabrica: Total AI Crawler Blockade Behind Cloudflare's JS Challenge

Every AI crawler tested — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, ChatGPT-User, Bytespider, Applebot-Extended, anthropic-ai, and Perplexity-User — receives a 403 Forbidden from Cloudflare's JS challenge page. The browser baseline returns 200 with 823KB of content, confirming the site is fully rendered for humans but completely invisible to machine crawlers. No AI model can read a single word of content directly from the live domain.

Crawler Access

The robots.txt at https://www.creativefabrica.com/robots.txt is itself behind the Cloudflare challenge wall — it returns a 403 with a noindex,nofollow meta tag and a Cloudflare Turnstile challenge page. This means no crawler can even read the crawl directives. The llms.txt file also 404s/403s. The sitemap (/sitemap_index.xml, /sitemap-1.xml) is equally blocked. The site runs on Cloudflare (NS: amy.ns.cloudflare.com, damon.ns.cloudflare.com) with IPs 104.18.10.141 and 104.18.11.141, using Cloudflare's Managed Challenge (Turnstile) — a configuration that gates all non-browser traffic behind a JavaScript proof-of-work.

Cold-Knowledge Gap

The LLM knows Creative Fabrica as a subscription marketplace for digital design assets (fonts, graphics, craft templates, embroidery files) founded in 2016 by Roemie and Anca Hillenwaard, with an "All Access" subscription model and a "Studio" design tool. It also recalls reputational signals from 2023-2024 — user complaints about auto-renewal policies, cancellation difficulty, and inconsistent asset quality. This prior knowledge is richer than what the live site exposes to crawlers (nothing), but it is entirely sourced from off-domain signals (reviews, social media, press) rather than the site's own content. The site's actual value proposition — "Everyone can be creative," "Over 1 million unique premium designs," "Personal, Commercial and POD use" — is invisible to models.

Schema Posture

The Wayback snapshot (December 2024) reveals the homepage contained a WebSite schema with SearchAction and an OnlineBusiness schema (type OnlineBusiness, not Organization) with full address details: Amsterdam, Netherlands, Chamber of Commerce number 70114412, VAT ID NL858147877B01. Social profiles link to Facebook, Pinterest, Instagram, and LinkedIn. The schema uses the slogan "Everyone can be creative." However, this schema is only visible in archived snapshots — the live site serves none of this to crawlers.

External Signals

DuckDuckGo returns zero search results for creativefabrica.com, site:creativefabrica.com, or any branded query — a complete deindexing from that search engine. The site runs on WordPress (evident from the Wayback snapshot showing Yoast SEO v12.3, xmlrpc.php, and wp-content paths), but the live WordPress installation is entirely hidden behind Cloudflare's challenge. The CDN (cdn.creativefabrica.com) serves static assets freely, but the main application is locked down. The studio.creativefabrica.com subdomain loads a JS-rendered app shell with zero visible text for crawlers.

Findings

1. All AI crawlers blocked by Cloudflare JS challenge High

   Every AI crawler tested (GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, ChatGPT-User, Bytespider, Applebot-Extended, anthropic-ai, Perplexity-User) receives a 403 Forbidden from Cloudflare's JS challenge page. The browser baseline returns 200 with content, confirming the site is fully rendered for humans but completely invisible to machine crawlers.

   What to change: Configure Cloudflare to allow AI crawlers by disabling the JS challenge for known bot user agents, or serve a static HTML version to bots.

2. Robots.txt inaccessible behind Cloudflare challenge High

   The robots.txt file at https://www.creativefabrica.com/robots.txt returns a 403 with a noindex,nofollow meta tag and a Cloudflare Turnstile challenge page. No crawler can read the crawl directives.

   What to change: Serve a static robots.txt file that is accessible to all crawlers, with appropriate directives for AI bots.

3. Sitemap files blocked by Cloudflare High

   Sitemap files at /sitemap_index.xml and /sitemap-1.xml return 403 errors, preventing crawlers from discovering the site's content structure.

   What to change: Ensure sitemap files are accessible to all crawlers by excluding them from Cloudflare's challenge rules.

4. llms.txt file not available Medium

   The llms.txt file at https://creativefabrica.com/llms.txt returns a 403, meaning AI models have no structured way to discover the site's content.

   What to change: Create and serve an llms.txt file that lists key pages and content for AI crawlers.

5. Complete deindexing from DuckDuckGo High

   DuckDuckGo returns zero search results for creativefabrica.com, site:creativefabrica.com, or any branded query, indicating the site is completely deindexed from that search engine.

   What to change: Investigate the cause of deindexing and ensure the site is accessible to search engine crawlers.

6. Schema markup only visible in archived snapshots High

   The Wayback snapshot shows WebSite and OnlineBusiness schema on the homepage, but the live site serves none of this to crawlers due to the Cloudflare blockade.

   What to change: Ensure schema markup is served in the initial HTML response and accessible to crawlers.

7. Studio subdomain renders empty JS shell for crawlers Medium

   The studio.creativefabrica.com subdomain loads a JS-rendered app shell with zero visible text for crawlers, making its content invisible to AI models.

   What to change: Implement server-side rendering or provide static HTML fallback for crawlers on the studio subdomain.

8. CDN serves static assets freely Low

   The CDN at cdn.creativefabrica.com serves static assets without challenge, but the main application is locked down.

What's working

• LLM has substantial prior knowledge of Creative Fabrica — The LLM knows Creative Fabrica as a subscription marketplace for digital design assets, with details about its founding, subscription model, and Studio tool. This prior knowledge is richer than what the live site exposes to crawlers.
• Archived homepage schema is well-structured — The Wayback snapshot shows WebSite schema with SearchAction and OnlineBusiness schema with full address details, social profiles, and slogan. This schema is correctly implemented but not served live.
• Site built on WordPress with Yoast SEO — The site runs on WordPress with Yoast SEO plugin, which provides a solid foundation for SEO and schema management if made accessible.