@playxbeasts · X

Flappy-minimal — 极简版 web flappy bird（100% AI 作品），纯娱乐项目，快来玩吧

@PremShandilya16 · X

Nobody decides to lose a friend. Life gets busy, everyone means to reach out, and then it's been three years. So I built ORBIT. You're the sun. Your friends are planets. Talk often and they orbit close. Go quiet and they drift away. Live, clickable 👉 Built with @flowstep_ai + Claude Code over MCP #FlowstepChallenge

@grimwalkerSM · X

CODES ON MY PROFILE

@ErwanCodes · X

I created Helios Control entirely with Grok 4.5. From concept to code, the whole project was generated, structured, and built with AI. A real test to see how far vibe coding can go in 2026. Live app 👇 Thank @SpaceXAI & @cursor_ai !

@OlesaAI · X

🚀 I'm building Solar Empire — a browser MMO strategy set in one seamless, living Solar System. And here's the twist: the entire prototype was pair-built with Claude Fable 5. The game. You run a private corporation. You start on Earth, where the geology is hidden — buying a hex IS the prospecting, so every claim is a gamble. Power plants feed mines, smelters feed factories, and a player-driven market ties it all together. Then you go to space. A rocket is assembled at your factory, rolls out to your spaceport pad, and launches — gravity-turn ascent, stage separation, the works. The Moon it flies to is actually moving: transfer windows change your fuel bill in real time, just like real orbital mechanics. The planets are real NASA/USGS maps. Weather is simulated once for everyone — the storm cutting your solar output is the same storm your rival sees from orbit. Abandoned bases decay into ruins that other players can scavenge, so the map never clogs up. The pro

@sun_dundun · X

做了一个游戏平台，全部使用Ai 来生成 #cursorcamp #game #Ai

@RanMarim4s · X

Vibe coding an entire game with AI agents sounds wild. Time to go test it out on 💎

HSR Warp Calculator — 《崩坏：星穹铁道》跃迁规划工具，计算保底进度、抽取概率和星琼需求

@spt4d · X

Deployed the 16 games built with Codex and GameBlocks. A few examples: Most of my time is actually spent shaping the player experience. Many ideas emerge through playing and exploration rather than from the initial design — the loop seems to be build → play → discover → improve. Is it possible to build an agent harness that supports this discovery loop — not just one that optimizes against a predefined spec? Full game list:

afitnerd · HN

Text adventure to learn skills and agents

nihey · HN

AlternateRealities.lol – Create any kind of poing and click story