Integrates advanced video perception, indexing, and timeline editing into Claude Code for programmatic media manipulation.
The VideoDB skill empowers Claude with the ability to perceive, index, and manipulate video and audio content across various sources including local files, URLs, live RTSP streams, and macOS desktop sessions. It allows users to build searchable visual and spoken indexes, generate subtitles or overlays, and perform programmatic timeline editing through a serverless infrastructure. This skill is ideal for developers building media-rich applications, monitoring live feeds for specific events, or automating video post-production workflows directly from their coding environment.
주요 기능
01Comprehensive media perception for video, audio, and live streams
0231,722 GitHub stars
03Visual and spoken indexing with timestamped search results
04Real-time desktop session capture and summarization for macOS
05Live stream (RTSP/RTMP) monitoring with AI-driven alerts
06Programmatic timeline editing for subtitles, overlays, and clipping
사용 사례
01Monitoring live security or broadcast feeds for specific visual triggers
02Automating video transcription and burnt-in subtitle generation
03Building a searchable video library with semantic and scene-based queries
