Proof of work

When I apply to a content/SEO/AI role, I run a live audit of the company's own site first — before anyone asks. Below are the actual findings, verified against the live sites on the dates shown. Credit where a site gets it right; gaps where the machines that power search and AI answers hit walls. Every claim here was checked directly — nothing in these audits is speculative.

Method note: only claims verifiable from an outside vantage are included. Where a check wasn't possible, it's marked as out of scope rather than guessed.

port.io — SEO/GEO Specialist application

2026-09-30

The citation layer exists, but the discovery layer is broken — Port's AI-readiness is genuinely ahead of most B2B SaaS, and it's undermining itself in robots.txt.

Direct fetches of robots.txt, /llms.txt, /llms-full.txt, sitemap.xml, homepage/docs/blog HTML. Every claim re-verified twice; a v1 misreading was caught and corrected in review.

  • ✓robots.txt explicitly welcomes AI crawlers — GPTBot, OAI-SearchBot, ChatGPT-User, PerplexityBot/User, ClaudeBot — plus Content-Signal: ai-train=yes. Most companies haven't made this decision yet.
  • ✓/llms.txt is substantial (~157KB) — the docs estate speaks to a machine audience.
  • ✓Blog posts carry real JSON-LD schema (Article, Organization, Person, WebPage, ImageObject) — verified on a sample post.
  • ✗robots.txt declares the wrong sitemaps: 'Sitemap: robots.txt' (itself — invalid) and 'Sitemap: llms.txt' (not sitemap-format), while the healthy ~100KB /sitemap.xml is declared nowhere.
  • ✗Homepage renders with zero JSON-LD — the highest-authority page carries no machine-readable identity.
  • ✗/blog and /docs are invisible to non-JS crawlers (client-side rendering) — exactly the crawlers that power AI answer engines.
  • ✗llms-full.txt is dead (404) despite being referenced — a discovery promise the site doesn't keep.

tailorbrands.com — SEO Associate application

2026-09-30

Technical floor genuinely clean — most sites get this wrong — but the AI-retrieval layer is missing entirely.

Direct fetches of robots.txt, /sitemap_index.xml, page- and post-sitemaps, /, /llc-formation/, /llms.txt, /llms-full.txt. Schema not scanned and deliberately not assessed (not verifiable from an outside vantage).

  • ✓robots.txt is minimal and correct — standard disallows only; GPTBot, ClaudeBot, PerplexityBot all crawl freely.
  • ✓XML sitemap properly declared and valid (Yoast index, page entries lastmod-dated the day of the scan).
  • ✗/llms.txt and /llms-full.txt both 404 — the site's crawl door is open to AI engines, but there's no retrieval layer waiting for them.
  • ✗Deliverable included: a ready-to-ship llms.txt draft built from verified sitemap URLs.

Want one of these for your site? That's the point — get in touch.