videodb

Ingest, index and analyze video files, RTSP/RTMP streams, and desktop captures into searchable timelines with subtitles and exports.

Updated Mar 1, 2026
One-click install
npx skills add https://github.com/derekhu0002/ai4pb-orchestrator --skill videodb-derekhu0002
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: videodb
Source: https://github.com/derekhu0002/ai4pb-orchestrator/tree/main/skills/videodb
Command: npx skills add https://github.com/derekhu0002/ai4pb-orchestrator --skill videodb-derekhu0002

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Video content is scattered across files and streams, making discovery, auditing, and reuse difficult. This Skill provides end-to-end ingestion, indexing, search, and AI-assisted processing to turn raw videos and live streams into searchable memory and actionable outputs.

Core Features & Use Cases

  • Ingest local files, URLs, RTSP/RTMP streams, and desktop captures for unified processing.
  • Index spoken words, visual scenes, and audio to enable semantic search and rapid retrieval of moments.
  • Assemble non-destructive timelines, generate on-demand streams, subtitles, and exports for reviews and collaborations.
  • Real-world use cases include product demos, meetings, live events, and monitoring pipelines that demand instant search and recaps.

Quick Start

Upload a video to your collection, index its spoken words and scenes, and generate a searchable, streamable result.

Frequently Asked Questions about videodb

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I index spoken words and visual scenes in a video for semantic search?▼

Video ingestion and indexing processes local files, URLs, and live RTSP/RTMP streams to extract spoken words and visual scenes. This enables semantic search across video content, turning raw footage into searchable memory for quick retrieval of specific moments.

Can I process live RTMP streams and desktop capture for real-time video analysis?▼

Yes, RTMP streams and desktop captures are supported for real-time video analysis. The skill ingests these live sources alongside local files into unified processing pipelines, enabling AI-powered transcription and scene indexing on streaming content.

What is non-destructive timeline editing for video processing pipelines?▼

Non-destructive timeline editing assembles video segments without altering original source files. It orchestrates on-demand streams, subtitles, and exports across end-to-end workflows, preserving raw footage for reviews and collaborations while generating actionable outputs.

Does this video indexing skill work with RTSP camera streams for monitoring pipelines?▼

Yes, RTSP camera streams are supported for monitoring pipelines. The skill ingests live RTSP feeds, indexes spoken words and visual scenes, and generates searchable recaps and on-demand streams for instant retrieval during live event monitoring.

How do I generate subtitles and exports from indexed video content?▼

Generate subtitles and exports by assembling non-destructive timelines from indexed video content. Once spoken words and scenes are indexed, the system produces on-demand streams and subtitle files for reviews, collaborations, and automated guidance.