Effortless and fast string similarity comparison using multiple algorithms, including bigrams, Levenshtein, Jaro-Winkler, and Ratcliff/Obershelp. Perfect for fuzzy search, NLP, and intelligent text matching.

Calculates similarity between two strings

A utility for analyzing text to find bigrams, trigrams, and other n-grams.

Measures how similar two strings are in terms of the number of common bigrams using Sørensen–Dice coefficient

Get n-grams from text

Letter pairs sorted by how frequently they appear in English words

Offline search with first-class CJK support, no segmentation dictionary.

plugin for nlp-compromise

Minimal implementations of a couple of classic text analysis tools (TF-IDF and cosine similarity)

The easiest way to get n-gram chunks from strings or token arrays!

opencode closed-loop usage coach — quota SENSE -> coaching DECIDE -> loop ACT + TUI integration

Portable multilingual term analysis, query planning, and highlighting for Byline CMS search providers

Alma memory layers: episodes, consolidated profile, budgeted recall, consolidation, structural erasure.

Detects similar strings using the Simon White's approximate string matching method, that computes the Sørensen–Dice similarity coefficient of sets of adjacent letter pairs.

Character-bigram tokenizer for Japanese full-text search, with an Orama adapter

RSS/Atom feed subscription, scheduled fetching, deduplication, LLM-powered summarization, and daily Markdown digests — as a dsh (DeepSeek Harness) bundle.

TypeScript port of SymSpell — symmetric delete spelling correction algorithm

Local embedding runtime (local/Ollama/remote) with a zero-download hash fallback

Browser/Worker binding for the corvid embedded database — the engine compiled to wasm32: idiomatic synchronous OOP (in-memory) plus async OPFS persistence via a dedicated Worker

Multi-granularity long-term memory engine: storage, retrieval channels, evolution.