The YouTube Transcript MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your YouTube Transcript account. It provides structured and secure access to video captions, stored transcripts, processing statuses, available languages, and playlist or channel video details, so your agent can retrieve captions, monitor transcription work, search saved transcripts, inspect language options, and find videos from playlists or channels on your behalf.
- Single video transcription: Have your agent fetch captions for a YouTube video and save the resulting transcript to your account.
- Batch caption retrieval: Direct your agent to retrieve and store captions for a group of up to 10 YouTube videos.
- Transcription status checks: Let the agent check whether individual or batch transcription work is processing, complete, failed, or waiting for speech recognition confirmation.
- Stored transcript access: Instruct your agent to list, search, and read transcripts already saved in your account without starting new transcription work.
- Language and video discovery: Have your agent review stored and translation languages for a transcript, or find video details from a YouTube playlist or channel.