home / mcp / youtube transcript mcp server
Provides MCP access to YouTube transcripts, including extraction, search, and time-based summaries for multiple videos.
Configuration
View docs{
"mcpServers": {
"alex2zimmermann-ux-yt-transcript-mcp": {
"url": "https://mcp.example.com/youtube-transcript",
"headers": {
"YT_MCP_MODE": "standalone",
"YT_MCP_TRANSPORT": "stdio",
"YT_MCP_BACKEND_URL": "http://localhost:8300",
"YT_MCP_CACHE_MAX_SIZE": "100",
"YT_MCP_BACKEND_API_KEY": "YOUR_API_KEY",
"YT_MCP_CACHE_TTL_SECONDS": "3600",
"YT_MCP_RATE_LIMIT_PER_MINUTE": "30"
}
}
}
}You can run a lightweight MCP server that creates, searches, and analyzes YouTube transcripts, enabling you to fetch full transcripts, search for keywords with context, and obtain time-based summaries across multiple videos. It integrates with MCP clients like Claude Desktop and Cursor, making it easy to incorporate YouTube transcripts into your workflows.
You will run the MCP server locally and connect an MCP client to it. Start the server in stdio mode, then in your MCP client configure the server using the provided command and environment settings. Once connected, you can request a full transcript for a YouTube video, search the transcript for terms with surrounding context, obtain a time-chunked transcript for analysis, or process multiple videos in a batch.
Prerequisites you need before installing include Python and pip on your system.
pip install yt-transcript-mcpIf you prefer installing from source, clone the repository and install the package locally.
git clone https://github.com/alex2zimmermann-ux/yt-transcript-mcp
cd yt-transcript-mcp
pip install .Configure the MCP server in your client to run in standalone mode by specifying the executable and environment settings. The server exposes an MCP entry point that accepts requests from your MCP client.
# Example configuration snippet for a client that supports MCP servers
{
"mcpServers": {
"youtube-transcript": {
"command": "yt-transcript-mcp",
"env": {
"YT_MCP_MODE": "standalone"
}
}
}
}This server provides a set of functions to interact with YouTube transcripts.
Get Transcript: obtain the full transcript from a YouTube video.
Search Transcript: locate keywords within a transcript with surrounding context.
Transcript Summary: receive a time-based breakdown of the transcript for easier analysis.
Batch Processing: process up to ten videos in a single operation.
Retrieve the transcript for a YouTube video with optional language and format.
Search a video transcript for a term with optional language and surrounding context segments.
Return a time-chunked transcript suitable for analysis.
Process multiple YouTube videos in a single operation, up to ten URLs.