"author:Gabrielc9951" matching MCP servers:

Turns YouTube videos into a persistent, compounding wiki knowledge base by extracting transcripts plus scene-change frames described by vision models and merging them into linked entity, topic, and concept pages. Enables natural-language questions over that compiled knowledge through local full-text search plus agent reasoning, with results returned to MCP clients like Claude or Copilot for citing.