Anil-matcha/vox-ai-motion-graphics-generator

What it solves

Creating high-quality, "Vox-style" paper-collage explainer videos is usually a slow, fragmented process requiring separate tools for scripting, image generation, animation, voice-over, music, and editing. This project automates that entire workflow into a single pipeline, ensuring consistent visual aesthetics and professional narrative structure without hours of manual labor.

How it works

The project functions as an automated agent that takes a topic, video, or photo and runs it through a structured pipeline. It first generates a narrative beat map and lets you pick a visual theme. Once approved, it creates collage-style posters for each beat, animates them using video models, adds narration and background music, and finally stitches everything together with captions using FFmpeg. You only need to interact with two human approval gates: the story beat map and the visual theme selection.

Who it’s for

This tool is designed for creators, video editors, and developers who want to produce editorial-style explainer content quickly. It is specifically built to be used by coding agents (like Claude Code or Codex), allowing users to generate complex videos simply by asking an AI assistant to create a "Vox-style collage video."

Highlights

• Three input modes: B-roll (topic to film), A-roll (talking-head to collage), and C-roll (product photo to anchored collage).
• Authentic collage aesthetic featuring torn paper, halftone dots, and bold headlines.
• Story-first approach that uses narrative arcs like "problem-agitate-solution" to guide the video structure.
• Integrated style bake-off to preview and select visual themes before final rendering.
• Fully automated assembly including narration, music ducking, and burned-in captions.
• Agent-native design that can be executed directly by coding assistants.