Skip to content
JevDirectory.orgTraffic count unavailable. View site traffic
CommunityJev Demos & Experiments

Coffee Under Fire

A React and Three.js coffee-delivery arena shooter whose enemies score their own tactics locally, with a decision record explaining why Jev was retired from that job.

See highlightsOpen github.com
Stars
0
Language
TypeScript
License
MIT
Published by
Community
Author
joaoh82
Last verified
Oct 2026

About this project

A React + Three.js coffee-delivery arena shooter. Every enemy plans its own tactics from what it can see, hear and remember, using a local utility scorer over legal actions; deterministic TypeScript executes the action. Read COFFEEUNDERFIREGDD.md for the design. Requires Node 24.x. Install with npm ci, then start the game and its local server. No API keys or external services are needed: Build before tests so the production startup test exercises the built application.

Quoted from the project's README. Read the full README.

Tagscommunityawesomejevcatalog-import
Works forJev for ClassificationJev for Routing
Used inJev for Software & DevTools
Languages
TypeScript

Highlights

  • Enemies plan from what they can see, hear and remember, scored by a local utility scorer over legal actions.
  • Jev was retired: docs/decisions/local-tactics.md explains the reason and the last version is tagged jev-concept locally.
  • Runs without keys or external services on Node 24.x, with npm run dev on loopback 5173 and the server on 8787.
  • Decision replays export seed, player inputs and applied decisions; replay() re-runs them within the same build only.

Watch out

Jev is not in the current build. Enemy tactics come from a local utility scorer over legal actions with deterministic TypeScript executing the result, npm run dev needs no API key, and docs/decisions/local-tactics.md explains the retirement while the last Jev version is tagged jev-concept locally. Earlier Jev-era playtest measurements survive in docs/decisions/playtest-08.md, but their measurement scripts were removed with the integration. Node 24.x is required, and the README is explicit that you must build before npm test so the production startup test exercises the built application. Development binds to 127.0.0.1 and the README warns against exposing it as a public game host because it bypasses production access controls; a trusted tunnel needs DEV_ALLOWED_HOSTS set to the exact assigned hostname and a Vite restart, and those settings add no authentication. Replay records are explicitly not promised to be compatible across builds, and the leaderboard is described as community-reported rather than anti-cheat verified. Not documented anywhere: whether the local scorer was benchmarked against the Jev version it replaced, or how its tactics quality was measured. 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.

More Jev resources like this

flue-jev-demo
11GitHub stars
Standalone Flue example that routes support chat through Cloudflare Workers AI, using gpt-4o-mini for chat and Jev for typed intent and urgency routing.
Jev Demos & Experiments#community#awesomejev#catalog-import
Communitymatthewp/flue-jev-demoTypeScript
ui-generator-instinct-jev
7GitHub stars
Demo that turns free-text UI descriptions into real shadcn components using only Jev typed Choice, Noul, and Score answers, with no generated code on the decision path.
Jev Demos & Experiments#community#jevbest#catalog-import
Communityjoevidev/ui-generator-instinct-jevTypeScript
jev-browser
4GitHub stars
TypeScript browser harness where an OpenRouter planner splits goals into subgoals, Jev picks each concrete action, OpenRouter fills text fields, and Playwright drives the page.
Jev Demos & Experiments#community#jevbest#catalog-import
Communityvinilana/jev-browserTypeScript
jev-lab
7GitHub stars
Two TypeScript monorepo products that keep the rules in the engine and ask Jev exactly one question: what should this person or this shogi side do next. Hundred runs a hundred NPCs with hunger, money, jobs, relationships and memories; Jev Shogi picks Gote's reply from engine-legal USI moves.
Jev Demos & Experiments#community#jevbest#catalog-import
Communityjammaru/jev-labTypeScript
otto
4GitHub stars
Open-source Electron computer-use agent for macOS and Windows that reads the apps you select through native accessibility with local OCR added for text targets, and asks Jev to choose concrete actions that it shows you before executing.
Jev Demos & Experiments#community#jevbest#catalog-import
CommunityNobleSpartan6/ottoTypeScript
harden-jev-decides
1GitHub stars
A decision board that sends one System One request describing eleven stream ideas, then lets code gate and weight Jev's answers into a single pick for what to build next.
Jev Demos & Experiments#community#jevbest#catalog-import
Communitytylerjharden/harden-jev-decidesTypeScript

From the Jev guides

Original write-ups on what this entry is for.

Per-Tick Decisions in Games
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
Back to all resources

From the Jev community

Posts from builders shipping with Jev right now.

Follow @typesafeai