Skip to content


Switch branches/tags

Unicode Emoji JSON Test status

This library provides a up-to-date version of emoji data from Unicode in JSON format, in a number of easily consumable file structures.

Check out muan/emojilib for emoji keyword pairings for extended searchability.


RGI only

This data does not contain minimally-qualified and unqualified emoji.

RGI: Recommended for General Interchange. A subset of emojis which is likely to be widely supported across multiple platforms.

Minimally-qualified or unqualified emoji zwj sequences may be handled in the same way as their fully-qualified forms; the choice is up to the implementation.

Full description can be found at

Skin tone variations

Emoji's skin tone variations are consolidated into one base entry, with a skin_tone_support flag on them.

This means one entry of 👋 represents its 5 variations– 👋🏻, 👋🏼, 👋🏽, 👋🏾, 👋🏿; while raw unicode data list them as individual emoji entries.



  "😀": {
    "name": "grinning face",
    "slug": "grinning_face",
    "group": "Smileys & Emotion",
    "emoji_version": "2.0",
    "unicode_version": "6.1",
    "skin_tone_support": false
  "👋": {
    "name": "waving hand",
    "slug": "waving_hand",
    "group": "People & Body",
    "emoji_version": "2.0",
    "unicode_version": "6.0",
    "skin_tone_support": true,
    "skin_tone_support_unicode_version": "8.0"


  "Smileys & Emotion": [
      "emoji": "😀",
      "skin_tone_support": false,
      "name": "grinning face",
      "slug": "grinning_face",
      "unicode_version": "6.1",
      "emoji_version": "2.0"




  "light_skin_tone": "🏻",
  "medium_light_skin_tone": "🏼",


  1. npm run download

Download the latest data dump from Update the version variable in this file when a new version is available. Experiment with a version by passing an argument for version number: npm run download 13.0.

  1. npm run build

Parse and format the downloaded data into different files for distribution. This script also generates stats.json for use in test. Update the parser if the content format from unicode data has changed.

  1. npm test

Run test that ensures the build data matches the count of emoji parsed from the data source.

A common People's category test failure after an Emoji version upgrade might be that there are new dual skin tone emoji unaccounted for. Verify this by reading through the change log and see if there are any emoji that can be modified with two skin tone modifiers and add them to generate-emoji-counts.js then run npm run build which will update stats.json; then run the test again. This isn't automated currently. See #3.

Unicode License Agreement