nlp‑corpus is an npm package that bundles ~100 000 diverse English sentences—from movies, lyrics, social media, legal documents, news, fiction, and more—into a ready‑to‑load dataset for NLP experimentation. It’s aimed at developers and researchers who need a noisy, multilingual‑flavored text source (including slang, typos, emojis and occasional offensive content) to test or train parsers, language models, and other natural‑language‑processing tools.