The project
WROITER is a diagnostic tool for identifying AI-typical writing patterns in text. It analyzes surface features — rhythm, vocabulary range, structural templates, signposting density — and returns a profile of what it found and why: every flag points at specific text and names the pattern in plain language.
Most detectors give you a number and stop there. WROITER shows the flags themselves, explains what each one means, and is explicit about where the method is unreliable. The method is public. The specification is open. The limitations page exists because the failure modes matter as much as the detection logic.
The goal is to give editors, reviewers, and anyone who cares about writing quality a tool they can actually reason about — not a black box that accuses and moves on.
Who built it
WROITER is a 3AM Energy project, built by Alex Rostovtsev (admin@wroiter.com).
Anti-bypass stance
The method is published so writers can improve and reviewers can make better decisions — not so anyone can reverse-engineer evasion tactics. If you are looking for ways to make AI output pass detection, you are in the wrong place.