What problem does it solve? AI-writing detector output is often misread as proof of AI authorship, which can lead to unfair academic, hiring, publication, or disciplinary decisions. This Skill explains what detector flags actually mean, separates real signals from plausible false positives, and prevents scores from being turned into definitive authorship claims. ## Core Features & Use Cases - Evidence Classification: Distinguishes deterministic detector hits, model-only editorial observations, user-supplied context, and unavailable evidence. - False-Positive Analysis: Considers genre, second-language writing, technical register, editing tools, deadline pressure, and the writer's baseline style before interpreting flags. - Consequential-Decision Guardrails: Refuses to convert scores into claims of cheating, fraud, or dishonesty, and suggests more probative evidence such as drafts, revision logs, and source history. - Use Case: A hiring manager receives a detector report flagging a candidate's writing sample. Use this Skill to get a careful interpretation of what the flags do and do not establish, plus a list of evidence that would materially reduce uncertainty. ## Quick Start Review these AI-writing detector flags for plausible false positives and explain what they do and do not establish about authorship.
