// the find
muan/emojilib
Emoji keyword library.
A plain JSON data package mapping each emoji to an array of searchable keywords (e.g. 😀 → grinning_face, face, smile, happy). It's for anyone building an emoji picker or search feature who needs keyword-to-emoji lookup without writing their own dataset. As of 3.x it's keywords only — pair it with the author's unicode-emoji-json for names, categories, and skin-tone metadata.
Does one thing and stays out of your way — it's a data file plus a .d.ts, not a framework. CI actually tests the dataset itself (scripts/test.js in the workflow), which catches malformed or missing entries on every PR, not just syntax errors. The maintenance tooling (improve.js, review.js, augment-en.js) implies an actual editorial process behind keyword quality rather than a one-time dump that rots.
The 2.x→3.x split means you now need two packages (this one plus unicode-emoji-json) and have to merge them yourself to reconstruct what a single require used to give you — more surface area to version-match and more boilerplate for a basic lookup. Keywords are English-only (dist/emoji-en-US.json), so there's no real i18n story despite the locale-suffixed filename suggesting one was planned. There's no matching/ranking logic included — you still have to write your own fuzzy search or scoring on top of the keyword arrays, so 'searchable' is aspirational until you build that part. README covers migration in detail but says almost nothing about how keywords are chosen or how review.js/improve.js actually work, so extending the dataset yourself means reading the scripts.