
from everything-openai-codex78
Tools for ingesting, indexing, searching, editing, and transcoding video/audio (files, streams, desktop capture) with timeline editing, transcripts, and alerts
VideoDB provides an agent-facing skill for perceiving, indexing, searching, and programmatically editing video and live streams. The skill enables ingesting files or RTSP streams, extracting spoken-word transcripts and visual indices, searching by semantic queries with timestamped evidence, and producing edited assets (subtitles, clips, overlays, transcodes). It's designed to offload heavy media tasks to a VideoDB service rather than local ffmpeg.
Use this skill when the user needs to: capture desktop sessions, transcode or reframe videos for social platforms, build searchable timelines, generate subtitles and translations, create clips from search results, or run real-time alerts on live RTSP streams. It's appropriate for media automation, moderation, and programmatic content production workflows.
Best for agents that can run Python snippets and call external services (Codex, Cursor, Gemini CLI) and for automation pipelines that need media processing as a service.
This skill has not been reviewed by our automated audit pipeline yet.