trmlabs.com — AI Site Grade

Analysis

TRM Labs — AI-Visibility Audit

The site has an llms.txt that is more detailed and current than what the cold LLM model knows about the company — the model still cites $130M in funding and 2018 founding, while the site discloses $220M in Series C funding and positions itself as an AI-era public safety operating system, not just a Chainalysis competitor.

Crawler Access

All major AI crawlers — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, OAI-SearchBot, ChatGPT-User, anthropic-ai, Applebot-Extended — receive a full 200 response with identical byte-size content (~231KB) to a browser baseline. The robots.txt explicitly allows every listed AI bot with no disallow rules. Only Bytespider (ByteDance/TikTok) gets a 403 from Cloudflare. The site runs on Cloudflare with HSTS preload and a permissive CSP (default-src *). No JS-rendering risk: all fetched pages return rich visible text on plain GET.

llms.txt and Structured Content

The /llms.txt file is a standout — 9KB with a curated key-facts table (193+ blockchains, 1.9B+ assets, 3.1B+ labeled addresses, 155+ risk configurations), categorized product links, solution pages, customer stories, and explicit citation guidance instructing LLMs to prefer the key-facts table over older blog-post numbers. This is an advanced implementation that few sites match. The sitemap indexes 1,291 URLs, indicating a large content surface. The FAQ page (/faq) is a rich Q&A resource with 764 words covering funding, mission, product descriptions at three technical levels, and customer names (Goldman Sachs, Circle, Anchorage Digital).

Cold-Knowledge Gap

The cold LLM describes TRM Labs as a "blockchain intelligence and risk management" company founded in 2018 with $130M in funding, compared to Chainalysis. The site tells a different story: $220M total funding (Series C), 150%+ annual revenue growth over five years, ~375 employees, and a repositioned narrative around "AI-era public safety infrastructure." The homepage leads with "Criminals scale with AI. Defenders scale with TRM" and features Orion (an AI investigator agent built into TRM Forensics) and Beacon Network (a real-time public-private intelligence-sharing network with Binance, Coinbase, Kraken, and Ripple as members). The cold model knows nothing about Orion, Beacon Network, or the 360 product suite (Compliance360, Investigation360, Seizure360, NatSec360, Supervision360).

Schema Posture

The homepage carries Organization, WebSite, and WebPage JSON-LD with correct sameAs (LinkedIn, Twitter), address, and a SearchAction target. The Forensics page uses WebApplication schema with BusinessAudience. However, the FAQ page has zero JSON-LD — no FAQPage schema despite having genuine Q&A content. The blog listing page also lacks any schema. The platform landing page (/blockchain-intelligence-platform) has a noindex robots meta tag, which is unusual for a primary product hub.

External Signals

The DNS TXT records reveal an unusually broad vendor stack: OpenAI domain verification (for ChatGPT plugin or custom GPT), HubSpot, LaunchDarkly, New Relic, Pardot, Notion, Linear, 1Password, Canva, Cursor, Jamf, and six Google site verifications. This signals deep API/partner integrations but also a fragmented tooling footprint. The Beacon Network page includes direct quotes from compliance officers at Binance, Coinbase, Kraken, and Ripple — strong third-party credibility signals that no cold model currently references.

Findings

1. Cold LLM knowledge omits $220M funding, Orion, and Beacon Network High

   The cold LLM describes TRM Labs with outdated $130M funding and misses key products like Orion AI investigator and Beacon Network, which are prominently featured on the site.

   What to change: Ensure the homepage and key product pages contain explicit, crawlable mentions of total funding ($220M), Orion, and Beacon Network in plain text, and update the llms.txt key-facts table to include these details.

2. FAQ page lacks FAQPage JSON-LD schema Medium

   The /faq page contains genuine Q&A content but has zero JSON-LD markup, missing an opportunity for rich results in AI and search contexts.

   What to change: Add FAQPage JSON-LD schema to the /faq page, marking up each question and answer pair.

3. Blog listing page lacks any structured data Medium

   The /resources/blog page has no schema markup, reducing its visibility in AI-generated summaries and search results.

   What to change: Add BlogPosting or CollectionPage schema to the blog listing page.

4. Platform landing page has noindex meta tag High

   The /blockchain-intelligence-platform page includes a noindex robots meta tag, which prevents it from appearing in search results and limits AI crawler indexing.

   What to change: Remove the noindex meta tag from the platform landing page to allow indexing.

5. Bytespider (ByteDance) blocked by Cloudflare Low

   The Bytespider crawler receives a 403 response, preventing TikTok/ByteDance AI products from indexing the site.

   What to change: If ByteDance AI visibility is desired, allow Bytespider in Cloudflare or robots.txt.

6. Zero external search results for TRM Labs queries High

   Multiple web searches for TRM Labs and its products returned zero results, indicating low external visibility and backlink profile.

   What to change: Investigate and improve off-site SEO, including backlinks, press mentions, and social signals.

7. No Reddit mentions found for TRM Labs Low

   A site:reddit.com search for TRM Labs returned zero results, indicating minimal community discussion.

   What to change: Consider engaging with relevant subreddits to build community presence.

What's working

• Advanced llms.txt with curated key-facts table — The /llms.txt file is 9KB with a detailed key-facts table, categorized links, and citation guidance, providing LLMs with accurate, up-to-date information.
• All major AI crawlers allowed with full access — GPTBot, ClaudeBot, PerplexityBot, Google-Extended, and others receive 200 responses with full content, ensuring broad AI indexing.
• Homepage has correct Organization and WebSite JSON-LD — The homepage includes Organization, WebSite, and WebPage JSON-LD with accurate sameAs, address, and SearchAction.
• Forensics page uses WebApplication schema — The /forensics page includes WebApplication schema with BusinessAudience, aiding AI understanding of the product.
• FAQ page provides rich Q&A content across three technical levels — The /faq page contains 764 words covering funding, mission, and product descriptions at beginner, intermediate, and expert levels.
• Beacon Network page includes quotes from major crypto exchanges — The Beacon Network page features direct quotes from compliance officers at Binance, Coinbase, Kraken, and Ripple, providing strong third-party credibility.
• Sitemap indexes 1,291 URLs — The sitemap contains 1,291 URLs, indicating a large content surface for crawlers to discover.
• Permissive CSP and no JavaScript rendering risk — The site uses a permissive CSP (default-src *) and all fetched pages return rich visible text on plain GET, avoiding JS-rendering issues for crawlers.