Slop*Detector

SUSPICIOUSLY VIBEY — slop score 56/100

The prose carries two em dashes doing parenthetical work ("not the Prague scholar who owned it, not the dealer who named it"), which is the punctuation habit most tied to generated writing, and it shows up right alongside a hedging filler line, "It is worth noting how lightly the book moved." The interface backs this up: lucide icons and radix primitives sit untouched at their defaults, and the page repeats its subscribe pitch four separate times, in the header, an inline block, a footer banner, and a popup modal that interrupts the comments section. None of these alone would convict a writer, but stacked together, with a mirrored negation structure bookending the essay, they point to a template someone filled in rather than a page built sentence by sentence. A 56 lands this in the suspicious range rather than confirmed slop; the research into the book itself seems solid and specific, so this reads like a human draft run through an AI pass and dropped into a stock Substack-style scaffold rather than a fully generated site.

Findings

• Pill tags — Little fully-rounded pill badges sprinkled across the page ("New", "Beta", category chips) are a default garnish of vibe-coded 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.
• Glassmorphism — Frosted-glass backdrop-blur panels are a recurring generated-UI decoration.
• Subscribe modal popup — A generic Substack-style modal interrupts reading with a subscribe form, duplicating the subscribe CTA already present in the header and in the footer banner — this is template scaffolding rather than a deliberate design choice.
• Redundant subscribe CTAs — The same subscribe prompt appears three times (header button, inline 'Subscribe to A Study in 221B' block, footer orange banner, and again in the popup modal), which is template repetition rather than purposeful design.
• Stock CTA banner — The bright orange 'Ready for more?' banner with an email field and Subscribe button at the bottom is a generic conversion-template close, not a bespoke sign-off tied to the piece's content.
• Em dashes — Em dashes are the single most reliable punctuation tell of LLM copy. Human marketing copy almost never uses them; models reach for them constantly.
• Hedging & filler-importance phrases — "It’s worth noting", "X is crucial", "it’s important to understand" — asserting importance instead of demonstrating it.
• Hollow superlatives — "Revolutionary", "world-class", "lightning-fast", "AI-powered" — claims with no content behind them.
• Em dash usage — The style guide explicitly bans em dashes under any circumstance; this piece uses them twice for parenthetical asides, which is the single punctuation tell most associated with generated prose.
• Hedging filler opener — "It is worth noting that/how" is a flagged hedging-filler construction that announces a point instead of just making it, and the guide specifically calls this phrase out.
• Mid-sentence dramatic colon — The colon here is used for rhetorical punch rather than to introduce a list, matching the banned 'dramatic colon' pattern.
• Mirrored negation-list structure — The same triadic 'not X, not Y, not Z' negation device is used twice (bookending the piece), echoing the guide's flagged 'dramatic negation list' rhythm even though each instance alone is mild.