3dicon README

The project's README from github.com/samyost1/3dicon. As one markdown file: README.md.

/3dicon

One prompt in, a looping animated 3D icon out (with real transparency) an animated WebP you can drop straight into your app UI.

eight icons looping around the /3dicon wordmark

Install

/plugin marketplace add samyost1/3dicon
/plugin install 3dicon

One OpenRouter key covers the whole pipeline — the image model and the video model both run through it.

cd ~/.claude/skills/3dicon
pip install -r requirements.txt
cp .env.example .env

Put your key in .env:

OPENROUTER_API_KEY=sk-or-...

Get one at openrouter.ai/keys.

ffmpeg must be on PATH. The first run downloads a ~180MB matting model.

Use

Ask for it in plain words:

make an animated 3d fire icon using /3dicon

It generates one still, shows it, and waits for you to approve it before spending anything on motion. Then it proposes the motion and waits again.

How it works

flowchart LR
    P["your prompt"] --> S["still image<br/><sub>GPT Image</sub>"]
    S --> F["first frame"]
    S --> L["last frame"]
    F --> V["video model<br/><sub>Seedance</sub>"]
    L --> V
    V --> M["remove background<br/><sub>every frame</sub>"]
    M --> E["animated .webp<br/><sub>real alpha</sub>"]

The trick is in the middle: the same still is sent as both the first and the last frame, so the model returns to where it began and the loop closes with no visible seam.

The background is removed against a colour we chose ourselves, which means the original colours can be solved for exactly rather than guessed — that is what keeps soft edges soft instead of leaving a halo.

Licence

MIT. Icons you generate with it are yours, with no restrictions.

3dicon (samyost1/3dicon)

Searching Database…

562 ·GitHub
3dicon
这是做什么的？一个 Claude Code 技能：用一句大白话描述你想要的效果，就能生成一个循环播放的 3D 动态图标——带真实透明通道的动态 WebP，可以直接放进应用界面里用。
它是独立应用，还是运行在 Claude Code 里面？它是 Claude Code 的一个技能，不是独立应用——你直接用日常语言要一个图标，比如说“用 /3dicon 做一个 3D 火焰动态图标”，它就会帮你跑完整个流程。
我要怎么安装配置？先拿到一个 OpenRouter 密钥（整个流程用这一个密钥就够了），填进 .env，并确保 ffmpeg 已加入 PATH——然后直接让 Claude 给你生成图标就行。
第一次运行前有什么需要知道的？ffmpeg 必须在 PATH 里，而且第一次运行会下载一个约 180MB 的抠图模型。生成是在 OpenRouter 的模型上运行的，所以这不是一个完全本地的工具。
生成的图标可以用在我自己的项目里吗？可以——代码采用 MIT 许可证，README 里也说明用它生成的图标归你所有，没有任何使用限制。
需要很强的显卡吗，还是普通笔记本也能跑？普通配置就够了：图像和视频模型都是通过 OpenRouter 调用的，本地只需要 ffmpeg，再加上第一次运行时下载一个约 180MB 的抠图模型。
实际生成一个图标要花钱吗？这个技能本身是 MIT 许可、免费的，但图像和视频环节走的是 OpenRouter 的付费模型——所以它在生成动画花任何钱之前，都会先征求你的确认。
What are similar projects?sprite-gen, artcraft, storycast, onetake, CodeVideoModel, procedural-film, godogen, scroll-world
Which topics is it in?background-removal, icon-pack, image-to-video, icons, claude-skill, openrouter, text-to-image, claude-code-skill, agent-skill, ffmpeg, video-generation, 3d
Creator
samyost1
Size
7.6 MB
Stars
562
Updated
CommitSep 23
Age
25d
Commits · 90 Days