Run a de-AI-slop mission

Run a Brief de-AI-slop mission for <page or surface>. Before editing, read the current copy and the closest product, brand, customer-language, terminology, legal, and design sources. Write a short meaning lock: what this says, who it is for, why it matters, the next action, and the claims or terms we must preserve. Don't invent positioning, metrics, urgency, guarantees, or capabilities.

Edit the copy in focused passes: clarity, reader value, evidence, voice, terminology, density, and anti-slop. Replace stacked abstractions with concrete nouns and verbs. Remove generic praise, vague transformation claims, fake certainty, throat clearing, forced contrasts, repeated three-part lists, and headings that communicate no reader value. Keep technical detail that helps the reader decide or act. Don't make the writing artificially terse, cute, poetic, or aggressive.

Run a headings-and-actions scan. Read only the navigation labels, H1 through H3 headings, CTA labels, and essential trust labels in order. Check whether a new reader can tell who this is for, what the product helps them do, how the argument progresses, and what to do next. Fix the first unclear point in the hierarchy before polishing lower-priority sentences.

Verify the edited copy in context. Confirm the meaning lock still holds, every material claim is supported or flagged, required terminology and legal language remain intact, and the copy matches implemented behavior. Check desktop and narrow layouts for heading wraps, paragraph measure, CTA wrapping, overflow, and density. Summarize what became clearer, what was shortened, what was intentionally preserved, and any owner decision still needed.