Slop*Detector

CERTIFIED SLOP — slop score 75/100

The gpteng.co script embedded in the page is about as close to a signed confession as this line of work gets, since Lovable leaves that fingerprint on every export it produces. Pair that with untouched shadcn buttons, stock lucide icons, and the default Inter/Space-Grotesk font stack, and you've got a shell nobody bothered to restyle after generation. The copy has its own tell too, with a straight

Findings

• Builder fingerprint: Lovable — The site carries provenance artifacts of Lovable. This is conclusive: whatever the page looks like, it was generated.
• Gradient backgrounds / gradient text — Large gradient washes and gradient-clipped headline text (usually blue→purple) are the default decoration of generated UI.
• Untouched shadcn/Tailwind defaults — shadcn components, lucide icons, and Inter/Geist/Space-Grotesk font stacks left at their defaults mean the model picked the design, not a person. Instrument Serif as an accent font is a separate, independently corroborated tell.
• Real, non-stock content and photography — The site uses actual photos of coffee shops and cups (varying quality, lighting, framing) rather than AI illustrations or stock imagery, and the reviews are specific, opinionated, and reference real local details (Lipscomb University, Egg Drip, FIU Biscayne Bay campus), which reads as human-written editorial content rather than generated filler.
• Minor stat/counter row — A small stats callout ('53 shops', '7 countries') appears near the section header. This is a low-confidence, unconfirmed tell per the guide, but worth noting as a mild templated touch.
• Repeated 'View Details' button pattern across grid — Every card in the shop grid uses an identical outlined button with the same icon and label, which is a mild templated-grid signal, though the underlying content (photos, reviews) is bespoke and not merely icon+subheader+blurb filler.
• "It’s not X. It’s Y." reframe — The contrast-reframe construction ("It’s not X. It’s Y.", "X isn’t just Y, it’s Z") is the classic LLM rhetorical move. A dead giveaway.
• AI vocabulary — Words statistically overrepresented in generated text: "seamless", "leverage", "unlock", "empower", "elevate", "robust" and friends.
• AI vocabulary — The word 'Elevate' appears in the mission statement, and it's one of the words flagged as overrepresented in generated text; everything else in the copy is otherwise specific and idiosyncratic, so this reads like a stray marketing-speak holdover rather than a generated pattern.