Skip to content

rea-video

Type: SkillCategory: Agentic video production★ 7864 (GitHub, when we checked)

rea-video is an agent skill in our Agentic video production category. JuneYaooo's Apache-2.0 Claude Code/Codex tool that breaks a video down shot by shot (filmed, AI-generated, avatar, 3D, code-built or screen recording) using frame cadence, noise, metadata and audio analysis, then writes a remake plan per shot (lighting setups, prompts and model picks, Remotion approach), timed captions and sound map. Outputs an HTML/PDF report plus a layered VIDEO.md for agents. Offline tools mode is free; optional Gemini/Doubao vision. For learning technique, not reusing others' footage.

Where it fits: Remotion.

Listing isn’t endorsement. This is our summary of the project’s own docs. Check the licence and requirements at the source before you build on it.

Lab posts that cover rea-video

Related Lab posts

Similar skills & repos