Process and Analyze Video and Live Streams

VideoDB provides perception, memory, and actions for video, live streams, and desktop sessions across five capability areas: (1) Desktop Perception - start/stop a desktop session capturing screen, mic, and system audio, stream live context, store episodic session memory, run real-time alerts on what's spoken or on screen, and produce session summaries with a searchable timeline and playable evidence links; (2) Video ingest + stream - ingest a file or URL and return a playable web stream link, transcoding and normalizing codec, bitrate, fps, resolution, and aspect ratio; (3) Index + search - build visual, spoken, and keyword indexes, search and return exact moments with timestamps and playable evidence, and auto-create clips from search results; (4) Timeline editing + generation - generate/translate/burn-in subtitles, add text/image/branding overlays and motion captions, add background music/voiceover/dubbing, and compose/export via timeline operations; (5) Live streams (RTSP) + monitoring - connect RTSP/live feeds, run real-time visual and spoken understanding, and emit events/alerts for monitoring workflows.

Use it when you need video or audio perception, indexing, search, or timeline editing from files, URLs, desktop sessions, or live streams, when the task involves timestamps, searchable evidence, subtitles, clips, overlays, or real-time monitoring alerts, or when you want one workflow spanning ingestion, understanding, retrieval, and media actions. Canonical use cases: "Start desktop capture and alert when a password field appears," "Index this folder and find every scene with people, return timestamps," "Connect this RTSP URL and alert when a person enters the zone."

Common inputs are a local file path, public URL, or RTSP URL; a desktop capture request (start/stop/summarize); or an operation spec (context request, transcode spec, index spec, search query, clip ranges, timeline edits, alert rules). Common outputs are a stream URL, search results with timestamps and evidence links, generated assets (subtitles, audio, images, clips), event/alert payloads for live streams, and desktop session summaries with memory entries.

Python usage connects via from dotenv import load_dotenv; load_dotenv(".env"); import videodb; conn = videodb.connect(), reading VIDEO_DB_API_KEY from the environment or a project .env file - connect() raises AuthenticationError if the key is missing. Setup installs via pip install "videodb[capture]" python-dotenv (or without the capture extra if that fails on Linux), with a free API key (50 free uploads, no credit card) from console.videodb.io. The skill explicitly never reads, writes, or handles the API key itself - the user always sets it directly.

Developers building video/audio search, desktop session monitoring, or live-stream alerting features who need ingestion, indexing, editing, and real-time monitoring in one workflow instead of stitching together separate transcoding, search, and alerting tools.