An algorithm trained on a vast corpus of text, including books, academic papers, Wikipedia, online forums, and other resources, amounting to hundreds of gigabytes of data and representing millions of books and articles, should be well-equipped to assess your writing accurately. In this case, I would say that it's assessment nailed it.