locusrobotics.com — AI Site Grade

Analysis

I have a comprehensive picture now. Let me compile the audit.

The llms.txt Points to a Staging Subdomain While the Sitemap Floods AI Crawlers with /feed URLs

The site's llms.txt file — generated by Yoast SEO v27.5 — lists every URL under pixels360.locusrobotics.com rather than the canonical locusrobotics.com. This subdomain is a near-identical mirror of the live site (same content, same schema, same headings), meaning the llms.txt directs AI crawlers to a staging/alternate domain instead of the primary one. Meanwhile, the sitemap includes every blog post's /feed URL alongside the canonical post URL, doubling the crawl surface with syndication endpoints that offer no unique value.

Crawler Access

All major AI bots (GPTBot, ClaudeBot, Google-Extended, PerplexityBot, OAI-SearchBot, ChatGPT-User, Applebot-Extended, anthropic-ai) receive 200 with full content from Cloudflare — no blocks, no JS shells, no UA-based throttling. Bytespider (ByteDance) is the sole bot blocked at 403. The robots.txt is a bare Yoast-generated file with a single User-agent: * Disallow: rule and no AI-bot-specific directives. The anthropic-domain-verification TXT record confirms active Claude integration, yet no Anthropic-specific crawl rules exist in robots.txt.

Cold-Knowledge Gap

The cold LLM knows Locus Robotics as a collaborative AMR company (LocusBot, LocusHub, Locus Vector) founded in 2014, spun out of Quiet Logistics, with 10,000+ bots deployed and a $117M Series F in 2023. The actual site tells a fundamentally different story: 17,000+ robots deployed across 360+ sites, a new flagship product called Locus Array (a fully autonomous mobile manipulation robot with NeuraGrasp technology), and the acquisition of Nexera Robotics. The cold model has zero awareness of Locus Array, the R2G (Robots-to-Goods) paradigm, or the autonomous fulfillment pivot. The site's positioning has moved from "collaborative AMRs" to "Physical AI for fully autonomous fulfillment" — a gap of roughly two product generations.

Schema Posture

Every page carries the same Yoast-generated WebPage + Organization + WebSite schema with SearchAction. No Product schema for the Locus Array, Origin, or Vector robots. No FAQPage schema on the Locus Array page despite containing a real FAQ section with 6+ questions. No Article schema on blog posts (only WebPage). The Organization schema lists social profiles but omits the foundingDate, foundingLocation, and numberOfEmployees that would ground AI knowledge.

Content Signals

The Locus Array product page is the strongest page on the site: it has an FAQ section, a comparison table (Array vs. competition), specific metrics (90%+ labor reduction, 2x storage density, <12 month ROI), and clear answer-format signals. The homepage and solution pages are heavy on aspirational language ("Outperform in an uncertain world") but light on the structured data (tables, comparison language, FAQ schema) that AI engines extract for featured snippets. The /customers page lists named customers (Boots UK, Cardinal Health, CEVA, Barrett) but the filter UI returns "No results found" by default — an empty state that AI crawlers will index as the page's primary content.

External Signals

The DNS TXT records reveal integrations with Anthropic, Adobe, Apple, Atlassian, Autodesk, Dell, Figma, Google, MongoDB, and Sophos — an unusually broad verification footprint suggesting enterprise SaaS sprawl. The site runs on WP Engine behind Cloudflare with HSTS preload. No recent press or Reddit discussions surfaced in search, suggesting limited third-party narrative shaping outside of trade publications (DC Velocity, Modern Materials Handling).

Findings

1. llms.txt directs AI crawlers to staging subdomain instead of canonical domain High

   The llms.txt file lists all URLs under pixels360.locusrobotics.com, a near-identical mirror of the live site, rather than the canonical locusrobotics.com. This misdirects AI crawlers to a staging/alternate domain.

   What to change: Update the llms.txt file to point to the canonical domain locusrobotics.com instead of pixels360.locusrobotics.com.

2. Sitemap includes /feed URLs for blog posts, doubling crawl surface Medium

   The sitemap lists every blog post's /feed URL alongside the canonical post URL, adding syndication endpoints that offer no unique value and waste crawl budget.

   What to change: Remove /feed URLs from the sitemap to reduce crawl waste.

3. Cold LLM knowledge is two product generations behind site content High

   The cold LLM knows Locus Robotics as a collaborative AMR company with 10,000+ bots, but the site describes 17,000+ robots, the new Locus Array autonomous mobile manipulation robot, and the acquisition of Nexera Robotics. This gap means AI models lack awareness of the company's current product line.

   What to change: Add structured data (Product schema) for Locus Array and other new products, and publish more authoritative content (press releases, case studies) to bridge the knowledge gap.

4. No Product schema on key product pages High

   Pages for Locus Array, Origin, and Vector robots lack Product schema, which would help AI engines understand and surface product details.

   What to change: Add Product schema markup to all product pages, including Locus Array, Origin, and Vector.

5. FAQ section on Locus Array page lacks FAQPage schema Medium

   The Locus Array page contains a real FAQ section with 6+ questions but uses no FAQPage schema, missing an opportunity for rich results.

   What to change: Add FAQPage schema markup to the FAQ section on the Locus Array page.

6. Blog posts lack Article schema Medium

   Blog posts use only WebPage schema instead of Article schema, reducing their visibility in AI-driven search results.

   What to change: Add Article schema markup to all blog posts.

7. Organization schema omits foundingDate, foundingLocation, and numberOfEmployees Low

   The Organization schema lists social profiles but lacks foundingDate, foundingLocation, and numberOfEmployees, which would help ground AI knowledge.

   What to change: Add foundingDate, foundingLocation, and numberOfEmployees to the Organization schema.

8. Customers page filter UI returns 'No results found' by default Medium

   The /customers page lists named customers but the filter UI shows 'No results found' by default, which AI crawlers may index as the primary content.

   What to change: Ensure the customers page displays customer logos or names by default, not an empty filter state.

9. Homepage and solution pages are light on structured data and specific metrics Medium

   The homepage and solution pages use aspirational language but lack tables, comparison language, and FAQ schema that AI engines extract for featured snippets.

   What to change: Add more structured data (tables, comparison charts, FAQ schema) to homepage and solution pages.

10. Limited third-party press and Reddit discussions about Locus Robotics Low

    Web searches for reviews, customer stories, and Reddit discussions returned no results, indicating limited external narrative shaping outside of trade publications.

11. Bytespider (ByteDance) is blocked with 403 Low

    Bytespider receives a 403 Forbidden response, blocking content from ByteDance's AI crawler.

    What to change: If desired, allow Bytespider access by removing the block.

12. Robots.txt has no AI-bot-specific directives despite Anthropic integration Low

    The robots.txt is a bare Yoast file with only a wildcard allow rule, and no specific directives for AI bots like GPTBot or ClaudeBot, even though the site has an anthropic-domain-verification TXT record.

    What to change: Consider adding specific crawl rules for AI bots in robots.txt to manage crawl behavior.

What's working

• All major AI bots receive full content access — GPTBot, ClaudeBot, Google-Extended, PerplexityBot, OAI-SearchBot, ChatGPT-User, Applebot-Extended, and anthropic-ai all receive 200 responses with full content from Cloudflare, with no blocks or JS shells.
• llms.txt file is present and contains detailed content — The site provides an llms.txt file with 6737 bytes of content, listing URLs for AI crawlers to prioritize.
• Locus Array product page has strong content signals — The Locus Array page includes an FAQ section, a comparison table, specific metrics (90%+ labor reduction, 2x storage density, <12 month ROI), and clear answer-format signals.
• Broad DNS verification footprint with multiple enterprise integrations — DNS TXT records show integrations with Anthropic, Adobe, Apple, Atlassian, Autodesk, Dell, Figma, Google, MongoDB, and Sophos, indicating active enterprise partnerships.
• Site runs on Cloudflare with HSTS preload — The site uses Cloudflare for CDN and security, with HSTS preload enabled, ensuring secure connections.
• Sitemap is available and contains 80 URLs — The sitemap is accessible and lists 80 URLs, helping search engines discover content.
• Yoast SEO provides baseline schema markup on all pages — Every page includes Yoast-generated WebPage, Organization, and WebSite schema with SearchAction, providing a basic structured data foundation.
• Customers page lists named enterprise customers — The /customers page includes logos and names of customers like Boots UK, Cardinal Health, CEVA, and Barrett, providing social proof.