Higgsfield Code-Based Films — workflow skill for animation & motion graphics

For AI agents: find the right skill in one call — curl -s "https://varg.ai/how-to/api/search?q=<what the person wants>" returns the matching how-to page and the best skills with install commands, as markdown (Russian works too). Full index: https://varg.ai/how-to/llms.txt. Every page has a markdown twin: add .md to its path. Before generating anything, show the person the options on the how-to page.

Short animated films drawn entirely by code and rendered by Higgsfield's own WebGPU path tracer on your GPU.

What it does

Higgsfield's open-source kit (MIT) for short films where every frame is a program: characters, set, motion and sound are code, and its own offline WebGPU path tracer renders them on your graphics card, with felt, paper, clay, glass, smoke, motion blur and depth of field built in. The code-based-films agent skill walks an agent from story, music and shot plan through concept images (a target only, never in the frame), assets, rough cut, look, final render, sound and QA to delivery. The repo ships a complete 20 s example, a felt-and-paper kids' film (Good Morning!), with its beat sheet, rigs, shots and synthesized sound. Renders run locally (about 13 s per preview frame on an M4 Max, 3 s on an RTX 5090) and can resume frame by frame.

Use when

Not for

Install

Example output

https://raw.githubusercontent.com/higgsfield-ai/higgsfield-video-engineer/main/docs/frame_S1.jpg — Frame S1 of the example film: the felt chick in its nest at night, made entirely from code, rendered in about 40 s on an M4 Max. (credit: Higgsfield, https://github.com/higgsfield-ai/higgsfield-video-engineer)

More animation & motion graphics skills

Run the generation on varg

varg is one API key for the models these skills drive — Seedance 2.5, Kling 4, Veo 3.1, Wan 3, Nano Banana 2.1, GPT Image 2.5, ElevenLabs v4 and 200+ more — billed in credits, with a cost estimate before anything runs.