Brain Dumps

By mid-2025, ~35% of new websites were AI-generated. But the data on whether everything 'looks the same' is more surprising than the meme. Here's what's real and what isn't.

1200x630 px, 1.91:1 ratio — that's the baseline. But the safe zone, the file-size ceiling, and the width/height attributes crawlers need are what most guides skip. Here's all of it.

You updated your og:image, you reshared the link, and it still shows the old card. Here's what's actually happening — and the exact steps to fix it.

A practical checklist for spotting AI-generated websites — the visual tells, the dead giveaways in the code, and how to confirm it by measuring the page instead of guessing.

Open Graph tags are the four lines in your that control how your link looks when someone shares it. Here's what each one does, the mistakes that break previews, and the minimum block every page needs.

AI-built sites are purple because of one default color in Tailwind UI that saturated AI training data — then fed back on itself. Here's the full chain, and how to break it.

What AI design slop is, why AI tools keep defaulting to the same purple-gradient, glass-card, low-contrast UI, what the data actually says about web homogenization, and how to catch it before it ships.

The internal link audit is one of the most valuable SEO tasks nobody does — because doing it by hand on a 100-page site takes a day. Here's the 50-line Python script + Claude prompt combo that runs it in 4 minutes.

Most meta-tag advice is either a decade out of date or a copy-paste list with no opinions. Here's what actually moves rankings and link previews in 2026 — and what to stop wasting time on.

You don't need Semrush, Ahrefs, and ContentKing combined. Here's the four-piece monitoring stack that catches the SEO bugs that matter — costs about €0 to run, alerts when something drifts, takes 15 minutes a week to review.

A canonical URL bug pointed half my pages at localhost:3000 in production for 11 days before I caught it. Then a 5-week recovery to get back to fully indexed. The timeline, the fixes, the GSC charts, and the lessons.

llms.txt is the AI-era robots.txt — a markdown file at your site root that tells LLMs what to read and what to skip. Here's what it actually does, how to write a real one, and the parts of the spec everyone gets wrong.