CoOp:Anything
게임을 검색해 협동 플레이 가능한 모드·단계·품질·커뮤니티 평점을 확인합니다
저장소 정보는 아직 수집하지 않았습니다.
아직 감지된 업데이트가 없습니다. GitHub 릴리스가 나오면 여기에 보입니다.
Interactive German Age of Empires IV counter learning app — standalone, responsive and auditable.
저장소 README에서 가져옴
README 전체를 불러오는 중입니다.
README 전문 보기 ↗Modes · How it works · Use it · Tech · Data
WAR ROOM turns raw Age of Empires IV unit data into something you can actually learn from. Instead of a static counter chart, it explains why a matchup tips the way it does — bonus damage, range, cost efficiency, terrain and micro — and lets you change the fight and watch the answer change with it.
It runs as a single HTML file: open it, host it anywhere, or use the live demo. No install, no account, no tracking.
1–4 / Enter, search: ⌘/Ctrl K), focus-managed, mobile-ready| ⚔️ Counter Lab | Pick your unit and the enemy. Get a clear verdict (from clear counter to hard loss), the combat-power ratio on a log scale, a model-confidence %, the reasons behind it and the full calculation. Ranked counter suggestions update as you tweak the combat variables. |
| 🔎 Unit Explorer | Browse and filter every unit. Each dossier shows stats, weapons, strong against / weak against lists and concrete usage tips. |
| ⏱️ Drills | The enemy is massing something — pick the best affordable answer in 12 seconds. Your streak and progress are saved locally. |
| 🧭 Fight IQ | Six practical combat rules (scout production, compare resources, control the contact, play the space, target by value, switch before you must) plus a 10-second pre-fight checklist. |
The app itself is bilingual — switch between 🇩🇪 Deutsch and 🇬🇧 English with the toggle in the top-right corner.
The matchup score is a learning model, not a frame-perfect combat simulator. It is built on Lanchester's square law in root form:
combat power = √(damage/s × hit points) × units × tactical factors
For each side, in the selected age, it computes
The two combat-power values form a ratio, rated on a log₂ scale (±0.3 = slight edge, ±0.78 = clear counter), paired with a model-confidence value and shown on an advantage scale. The "Show the maths" panel lays out every number.
Deliberate limits — civilization bonuses, activated abilities, landmark
effects, formation depth, real pathfinding and patch special cases can flip a
result. That's exactly why close matchups are marked as skill matchups rather
than pretending to be certain. All tuning constants are named and documented in
src/lib/matchup.config.js.
war-room.html from the latest release and open it by double-clicking.deploy/ for a ready-made CSP and step-by-step notes.React 19 + Vite 8, bundled into one file via vite-plugin-singlefile. Quality is
guarded by ESLint, Prettier, JSDoc type-checking, a Vitest suite (55 tests) and
GitHub Actions CI.
npm install
npm run dev # local dev server on http://127.0.0.1:5173
npm run build # production build → outputs/index.html (standalone)
npm run audit # lint · typecheck · test · build · standalone check
Updating the game data:
git clone --depth 1 https://github.com/aoe4world/data.git work/aoe4world-data
npm run generate:data
The generator writes every unit with its per-age stat variants and data-driven
combat flags into src/data/units.generated.js. It warns when listed
special units disappear, when a bonus group no longer matches any unit, or when
new class tokens appear — so balance patches surface instead of silently
breaking the logic.
Optional cookieless analytics (GoatCounter) are only embedded when you build
with VITE_GOATCOUNTER_CODE=<your-code>; without it the file loads no external
script at all.
The exact upstream commit and patch date are recorded in the header of
src/data/units.generated.js and shown in the methodology dialog.
Age of Empires IV and related marks are property of their respective owners. This is an independent, non-commercial fan learning project and is not affiliated with Microsoft or Relic Entertainment. Unit values are derived from the community dataset aoe4world/data; unit artwork and fonts load from their original sources and are not redistributed here.
The project's own source code is released under the MIT License. The disclaimer above covers third-party names, marks and data.