玉米

Ai设计从业者，日常分享一下自己接触到的好玩有趣的东西，分享一下经验

[玉米 @yumi_2333]

快来看看你的个人 AI 智能体 Muse 。在加入后的 48 小时内通过“设置”兑现我的邀请码，我们就能分别获得 10 亿个 Muse 词元。

邀请码：7KLU8O
https://t.co/oHocF9Ym9r

[玉米 @yumi_2333]

@Super_Nius

[玉米 @yumi_2333]

@mediastorm_en b站老粉来了

[Tim Pan @mediastorm_en]

My iPhone Duo video went viral, so I'm giving away 100 iPhones on X 😛

[玉米 @yumi_2333]

最近在写chat agent MCP缘由就是，sol太慢了，简单运维还得等半天

[玉米 @yumi_2333]

@BitcoinPulseX BTC

[玉米 @yumi_2333]

@ZacharyZhang Gemini的文笔太无敌了，比Claude强

[玉米 @yumi_2333]

@Jason_Young1231 @evermind 大佬这个真有用吗

[玉米 @yumi_2333]

@yetone GPTPro 我平常生图我完全不知道用的是哪个模型

[玉米 @yumi_2333]

@yetone 那个那个我能问一个问题不 我是图文工作者 我看到居然能选 image 模型吗 选了真有意义吗

[玉米 @yumi_2333]

购买 Claude 第一天 还没被封 日本家宽落地+内网阻断 保证 Claude 不连上国内网络 我将实时记录

[玉米 @yumi_2333]

听说其实是提示词的问题 gpt 提示词对了貌似也行

[zero @twoclipping]

opus 5.5 is f*cking cracked at motion design

this entire video is code, 0 after effects

im open sourcing the prompt template for these motion designs

steal it to recreate these ↓

<inputs>
Ask me for: 8 to 12 UI states I want the shape to become (e.g. button, loader, player, slider, toggle, tabs, chart, command palette, toast), pure black and white or one accent color, and a royalty-free song around 120 BPM (e.g. Mixkit, free for commercial use).
</inputs>

<direction>
Dribbble-level UI motion. One shape, never cut: every state is the same element morphing its size, radius and color while its content swaps with a short blur. A cursor drives every change with real clicks and drags. Light warm-gray canvas, black and white components, one clean UI font (Geist). Springs everywhere, a tiny overshoot at most. The camera zooms so each state fills the frame. The last frame is the first frame, so it loops.
Banned: bouncy easing, particle bursts, glows, gradients on UI chrome, mismatched icon strokes, dead time, anything that looks like a template.
</direction>

<structure>
120 BPM, 7 bars, something happens on every beat.
Button → loader → check → dynamic island → music player with a play/pause morph → scrub the progress bar → it becomes a volume slider that stretches when dragged past max → a toggle flips on the beat → the knob becomes a liquid tab indicator → the tabs open into a chart that draws itself, with a tooltip on hover → it collapses into ⌘K → type to filter → enter → toast → back to the button.
</structure>

<build>
1. One HTML file, square 1440x1440. Every style is computed from time inside seek(t): no CSS transitions, no timers, no state carried between frames.
2. Springs are closed-form step responses. A value that changes target many times is the sum of one spring per change, so it stays a pure function of time.
3. The tab indicator's two edges ride different springs, so the leading edge stretches ahead of the trailing one. Same trick for the toggle knob.
4. Drags are direct manipulation: while the cursor is held, the value is computed from its position. On release it springs back from wherever it was.
5. Analyze the song with numpy for the beat grid and start on a downbeat. Place every UI sound by its measured peak.
6. Render with Playwright: 4 subframes per frame, blended with ffmpeg tmix for motion blur at 60fps.
7. Render one frame per beat before the full render. Fix anything off the grid, cramped or hard to read.
</build>

<gotchas>
Never put will-change on anything the camera scales or the text renders blurry. Text that swaps inside a morphing container needs its own enter and exit timing or it overlaps. Make the last frame identical to the first, cursor position and speed included, or the loop stutters.
</gotchas>

<start>
Ask me for the inputs, then show me the state list on the beat grid before you write any code.
</start>

[玉米 @yumi_2333]

动效分享

[Aymen @aymenpursue]

Someone built a growing gallery of motion graphics made with Claude Opus 5.5, each shown alongside the prompt or skill that created it.

226 examples + the prompts behind them 👀

https://t.co/5eFiPHZ7As

[Simon聊AI @HanZhang415188]

https://t.co/oRUuTPdjs2

[花叔 @AlchainHust]

替你们试过了，我这个skill做短视频起号绝了。

晚上出门前随手花10分钟做的视频，已经3万+播放了。比我日常耗费精力做的视频中位数水平好多了。

[nicco @0xnicc0]

opus 5.5 just made this entire launch film from one sentence

no After Effects. no Figma. no editor. every frame is code

prompt → concept → storyboard → motion → sound → render

i'm sharing the exact prompt template, grab it and recreate it ↓

<inputs>
Ask me for: a one-word product name for the wordmark, one accent color, and a royalty-free track around 120 BPM (e.g. Mixkit, free for commercial use).
</inputs>

<direction>
Product-launch UI motion in one continuous take. One shape tells the whole story: a prompt bar becomes a button, the button becomes a progress ring, the ring becomes a preview window, the window fills the screen, then shrinks into a toast and back into the prompt bar. No cuts, no fades between scenes: objects transform. A cursor triggers every change with real clicks. Warm off-white canvas, black UI, one accent color, Inter for UI and the wordmark. Springs everywhere with only a slight overshoot. Zoom the camera so each moment fills the square.
Banned: crossfades, gradients on UI, particles, glows, bouncy easing, dead beats, anything that feels like a template.
</direction>

<structure>
120 BPM, 24 beats, something happens on every beat.
Type the request into the prompt bar → click send → the bar collapses into a black circle → a progress ring fills in four steps while four chips spring out (concept, storyboard, motion, sound) → the chips fly into a timeline and the circle becomes a preview window → a playhead sweeps the clips → inside the preview an orbit draws itself, a planet starts moving and the wordmark rises letter by letter from a mask line → the preview fills the screen → the letters drop away, the planet flies into a toast and becomes its checkmark → "Rendered in code · 60 fps" → the toast turns back into an empty prompt bar. Last frame = first frame.
</structure>

<build>
1. One HTML file with a 1440x1440 canvas. Draw everything inside seek(t): no CSS transitions, timers or state carried between frames.
2. Use closed-form spring step responses. When a value changes target several times, add one spring per change so it stays a pure function of time.
3. The main shape is one rounded rect whose width, height, radius, position and color are each a chain of springs.
4. Typing is derived from time: characters shown = floor(progress × length), with a caret that blinks only when idle.
5. Find the beat grid of the track, start on a downbeat and place every click, pop and chime on its measured peak.
6. Render with Playwright at 60 fps, then encode H.264 CRF 16 and loudnorm to -14 LUFS.
7. Render one still per beat first. Fix anything cramped, off-grid or hard to read before the full render.
</build>

<gotchas>
Text inside a morphing shape needs its own clip and separate enter/exit timing. Text swaps need their own fade so they never overlap. The cursor must end at the exact position it started, or the loop will stutter.
</gotchas>

<start>
Ask me for the inputs, then show me the beat map and 4 stills (prompt, progress, preview, toast) before writing the full film.
</start>

[玉米 @yumi_2333]

严肃学习

[卡尔的AI沃茨 @aiwarts]

超爆肝！今天把在几个月前我在影视飓风内部分享的AI工作流全公开！

00:15 找到适合你的Skill
09:28 把你的经验打包成Skill
20:34 把多个Skills组合成工作流
30:54 批量化定时化

看完之后你甚至可以手搓一个本地批量视频剪辑Skill，还能做出文案、海报、视频、网页全链路的内容生产工作流！！

[玉米 @yumi_2333]

@p4nthera_ 这个月也要试试 Claude 了 ChatGPT 5x 不够用了…

[玉米 @yumi_2333]

omg 这个 dots 出现的太是时候了 懒得搞其他的 bot 了

[玉米 @yumi_2333]

@iiiwuxxx 我这提交后显示了要3天才下来要那么久吗