videodb

Ingest, index, search, and stream video with Python and VideoDB.

86|21|Updated Feb 9, 2026

Search Tags:#indexing #ai #search #video #streaming #capture

Ready to run

npx skills add https://github.com/Jamkris/everything-gemini-code --skill videodb-jamkris

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: videodb
Source: https://github.com/Jamkris/everything-gemini-code/tree/main/skills/videodb
Command: npx skills add https://github.com/Jamkris/everything-gemini-code --skill videodb-jamkris

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Video content is hard to find and reuse across long recordings. This Skill provides perception, memory, and action pipelines to ingest, index, search, and stream video and live data.

Core Features & Use Cases

Ingest + Index: ingest files, URLs, and live streams; index spoken words and visual scenes for fast retrieval.
Search & Retrieve: semantic and keyword search with timestamps, clips, transcripts, and playable streams.
Timeline Editing & Streaming: non-destructive timeline composition with server-side transcoding and instant on-demand streams.
Capture & RTStream Workflows: desktop capture, RTSP/RTMP ingestion, and real-time AI pipelines for memory creation and alerts.

Quick Start

Install VideoDB, configure your API key, and run a quick workflow to ingest, index, and search video content.

Frequently Asked Questions about videodb

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

FAQPage Schema

How do I search video content for specific spoken words or visual scenes?▼

To search video content for specific spoken words or visual scenes, you can ingest files or live streams into a centralized memory that indexes spoken words and visual scenes. This enables fast retrieval using semantic and keyword search with timestamps, clips, and transcripts. Server-side processing handles the indexing and transcoding.

Can I ingest live RTSP or RTMP streams into a searchable video memory?▼

Yes, you can ingest live RTSP or RTMP streams into a searchable video memory. The pipeline supports RTSP/RTMP ingestion and desktop capture, routing live data into a centralized, indexed collection for real-time transcription, visual and audio indexing, and immediate retrieval.

What is the best way to compile video clips and highlights across long recordings?▼

The best way to compile video clips and highlights across long recordings is to use non-destructive timeline editing with server-side transcoding. This approach allows you to compose playable streams on-demand from indexed video segments, avoiding manual scrubbing and exporting.

Do I need a Python runtime and API key to process live video captures?▼

Yes, you need a Python runtime and a VideoDB API key to process live video captures. The Skill relies on this environment to execute ingestion, run indexing workflows, and communicate with server-side processing for transcription and transcoding.

How does real-time transcription and visual indexing work for streaming video?▼

Real-time transcription and visual indexing for streaming video work by capturing live data feeds and sending them to server-side processing. The system automatically indexes spoken words and visual scenes as they arrive, creating a searchable timeline of events that can trigger alerts or compile highlights.

Are there limitations when using server-side transcoding for on-demand video streaming?▼

Server-side transcoding for on-demand video streaming requires continuous API connectivity and depends on server-side processing capacity. While it enables instant stream compilation from non-destructive timelines, local processing is not supported, meaning all indexing and transcoding must occur on the remote server.