Vincentwei1021/video-shotcraft
AI video skill for Claude Code & Codex — cinematic product videos with Remotion: 152 shot recipe cards, 209 motion previews, a production-ready template
What it solves
video-shotcraft provides a structured system for AI agents to create professional, cinematic product promo videos. It eliminates the need for manual motion design by providing a library of pre-defined "shot recipe cards" and templates that an agent can use to storyboard, animate, and add sound design to a product's marketing materials.
How it works
The project functions as an "agent skill" that can be integrated into AI coding agents like Claude Code or Codex. It uses the Remotion framework to programmatically generate videos from React components. The agent follows a production methodology (capture, visual direction, storyboarding, and beat-syncing) and selects from a library of 157 shot recipe cards and 214 motion previews to build a film. It also includes a browser-based Motion Workbench for post-delivery editing and the ability to export projects to JianYing (CapCut CN).
Who it’s for
This tool is designed for developers and marketers who want to use AI agents to quickly produce high-quality product demos, launch videos, and brand films, specifically for web and desktop applications.
Highlights
- Extensive Motion Library: Includes 157 shot recipe cards and 214 motion previews with tuned timing and easing parameters.
- Production-Ready Template: Ships with the "Ink Press" template, a 36.2-second cinematic promo with 2.5D camera moves and pinned SFX.
- Agent-Native Workflow: Designed to be installed as a skill for Claude Code or Codex, allowing agents to handle the storyboarding and implementation.
- Post-Production Tools: Features a browser-based Motion Workbench for editing and a JianYing/CapCut CN project export for professional grading and re-timing.
- Sound Design System: Includes a curated library of 149 SFX across 16 scene categories to ensure beat-synced audio.
Related
- Project
- Project
- Project
- Project
- Project