1254 hand-tested AI prompts across image, video, and music — each with full context, variables, and a sample output. Content is authored in Traditional Chinese; the UI here is English.
電商產品旋轉最怕忽快忽慢+光影跳——這條明講 `no acceleration or dec…
破壞鏡頭的說服力靠碎片彈道與塵埃——`physics-accurate trajectori…
Kling 2.5 Turbo 能吃「一句連寫多鏡頭」——低角廣角→dolly back 退…
同款「一句流」災難模板換題材——生機械巨蛇破柏油、dolly back 穿煙、cut to …