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

/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.

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.

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.

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