14 Repos That Give AI Coding Assistants Design Sense, 3D Skills, and Video Editing
One developer with an AI assistant can now produce a marketing page, a Three.js scene, and a short film without leaving the editor, shrinking the team and skill set needed for digital content.
Over the past year, the author's saved repositories leaned away from source code and toward Skills, MCP servers, and plugins. The reason is that AI coding tools still struggle with aesthetics, modeling, motion, and editing. The list gathers 14 projects into three groups. One group supplies design knowledge, review, and animation components so Claude Code or Codex stops producing template-looking pages. Another connects AI to Blender, Three.js, and motion retargeting, covering the asset pipeline. The third treats video as code, using HTML, Remotion, and camera recipes to produce finished films.
The unifying thread is that front-end engineers can now own design, 3D, and video in one workflow. The author is blunt about the limit: these tools don't judge what good looks like, and the real advantage comes from wiring them into a project and knowing which steps still need human control.
The shift in saved repos from raw source to Skills, MCPs, and plugins shows that the AI coding ecosystem now treats design, modeling, and video as first-class problems rather than edge cases.
The author's framework groups tools by pipeline stage instead of by language, which suggests the next competitive edge for AI assistants may be how well they plug into existing creative software like Blender and Figma.
The closing argument that tools still need human judgment runs against the hype around autonomous agents, and it's worth weighing when deciding where to automate.