prose

Why we included this project

prose is a pure-Go library for the classic text analysis jobs: tokenization, sentence segmentation, part-of-speech tagging, and named-entity extraction, with no cgo and no calls to external services. That makes it a convenient fit for Go teams that want NLP running in-process instead of routing text to a Python service or an API. One detail that matters in real use is that every result carries a byte offset back to the original text, so you can verify exactly where a token or entity came from even when the input has multi-byte characters or odd whitespace. The library also lets you import just the pieces you need, so a binary that only does part-of-speech tagging skips the entity model entirely, and the components are safe to share across concurrent goroutines. Go developers building search indexes, readability scoring, or document preprocessing should find a clean, dependency-light base here.