The VideoDB MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your VideoDB account. It provides structured and secure access to your media collections, videos, audio files, and images, so your agent can organize collections, ingest media, find content, answer grounded questions, and manage media on your behalf.
- Media collection organization: Have your agent create collections, review available collections, and update their names, descriptions, or visibility.
- Video, audio, and image ingestion: Direct your agent to add media from a web address to the collection you choose.
- Semantic media search: Instruct your agent to search indexed media by meaning and return content relevant to your request.
- Grounded collection questions: Let the agent ask questions about a collection and check when each answer is ready.
- Media discovery and management: Have your agent list videos, audio files, or images, review their metadata, and rename selected media.