User repository on GitHub
Type anything. Get related emojis scored 0–3 with Jev.
- Forks
- 0
- Last pushed
- Updated Sep 2026
Repo stats from the GitHub API, cached Oct 2026.Homepage
CommunityJev Demos & Experiments
Jevmoji
Type up to 40 characters and get many related emojis scored by Jev, with category score traces and a keyword hits shortlist on a hosted Worker demo.
- Stars
- 6
- Language
- JavaScript
- License
- MIT
- Published by
- Community
- Author
- cheeaun
- Last verified
- Oct 2026
About this project
Under OUTPUT → score trace, categories list the emoji groups Jev scored. You may also see a row labeled hits. That row is not a Unicode emoji category. It is a shortlist the app builds in code by matching your text against emoji names and keywords (for example spiderman → spider / web). Those candidates are then scored like everything else. The Emoji→pages cell stays blank for hits because that column is for paging large catalog groups, not a keyword shortlist.
Quoted from the project's README. Read the full README.
Used inJev for Software & DevTools
- Languages
- JavaScript
Highlights
- Scores emoji groups per query and shows them under OUTPUT score trace beside the results.
- Builds a hits shortlist in code from emoji names and keywords, then scores it like other groups.
- Keeps the TypeSafe key server-side in the Worker, never in browser JS or config files.
- Rebuilds the catalog from Unicode plus emojibase keywords with a Worker-only generated JSON file.
Watch out
Hits is a keyword shortlist, not a Unicode category, and a missing hits row means local matching found nothing; the paging cell stays blank for shortlists. Copy reviewed against the repository README on 2026-10-04, not from running it.
Questions about this entry
Short answers drawn from the linked project or page.
That is the design: it returns many scored matches with per-group traces rather than one winner. Open the score trace to see which emoji groups Jev scored for your text.
Hits is a shortlist the app builds in code by matching your text against emoji names and keywords, then scores like everything else. The paging cell stays blank because that column pages large catalog groups, not a shortlist.
It never reaches the browser. Local runs read TYPESAFE_API_KEY from .dev.vars and production reads a wrangler secret server-side. Never put the key in client JS, .env, or the README.
Run the catalog script, pinned to Unicode 17.0.0 by default with older and newer version flags. Each entry keeps emoji, name, category, status, and keywords while hex, subgroup, and ids are omitted to keep the Worker bundle small.
npm install, write TYPESAFE_API_KEY into .dev.vars, then npm run dev, which is Vite plus the Worker with HMR, on http://127.0.0.1:8787. npm test runs the unit tests through node --test, npm run build writes to dist/, npm run deploy does a vite build plus wrangler deploy, npm run catalog rebuilds the emoji data, and npm run probe is a live pipeline probe for a query string that needs a key. The hosted version needs no install at all; it is a deployed Worker.
Five environment variables. TYPESAFE_API_KEY is required and server only. TYPESAFE_MODEL defaults to jev-latest. TYPESAFE_TIMEOUT_MS defaults to 60000 and is the SDK timeout per attempt. TYPESAFE_PRICE_PER_MTOK defaults to 0.042 and is for cost display only, not billing. EMOJI_PAGE_SIZE defaults to 80. The input itself is capped at 40 characters, which is the real limit on what you can ask for.
MIT for the project code, with the licence file referenced in the repository. TypeSafe and Jev themselves are proprietary and are explicitly not covered by that licence, so the model you call is not part of what you are licensed to reuse. The emoji data is generated rather than hand-edited, and the catalog script defaults to Unicode 17.0.0 because fonts often lag the newest release.
More Jev resources like this
23GitHub stars
Browser extension for Chrome and Firefox that scores X, LinkedIn, Reddit, and YouTube items on AI-writing tells and collapses posts over your threshold.
Jev Demos & Experiments#community#awesomejev#catalog-import
Community
adamnroman/slop-filterJavaScript
adamnroman/slop-filterJavaScript46GitHub stars
Open 150M encoder that reproduces the typed noul, choice, and score interface in one forward pass, trainable on a free GPU and runnable locally for nothing.
Jev Demos & Experiments#community#jevbest#catalog-import
32GitHub stars
Jev-protocol random baseline that answers noul, choice, and score questions from fast pseudorandom numbers without reading the prompt.
Jev Demos & Experiments#community#jevbest#catalog-import
Community
karminski/Jev-QuantumRust
karminski/Jev-QuantumRust30GitHub stars
Side-by-side game harness where Laya running locally and Jev over API answer the same typed questions for snake, kombat, runner, flappy, tetris, and human-versus-model matches.
Jev Demos & Experiments#community#jevbest#catalog-import
40GitHub stars
Python decision client exposing Choice, Noul, and Score primitives with isolated order-invariant scoring, a deliberative flywheel that crystallizes hard cases into fast paths, and local or hybrid deployment.
Jev Demos & Experiments#community#jevbest#catalog-import
Community
zhangcy122/OpenJevPython
zhangcy122/OpenJevPython23GitHub stars
Chrome extension that sends each X post to Jev with six typed questions in one call near the viewport, then renders a behavior chip under the post plus a session cost and latency panel.
Jev Demos & Experiments#community#jevbest#catalog-import
Community
oso95/x-scannerTypeScript
oso95/x-scannerTypeScriptFrom the Jev guides
Original write-ups on what this entry is for.
A game tick is a state plus a legal action list. How Clash Royale, Minecraft, and the earlier Mario, Pokémon, StarCraft, and drone loops ask Jev one step at a time.
Jev Demos & Experiments#gaming#agents#control
Read article
From the Jev community
Posts from builders shipping with Jev right now.