Home / MCP / Youtube MCP Server

Youtube MCP Server

Bridges YouTube data to AI assistants via MCP to download captions and generate video summaries.

python
Installation
Add the following to your MCP client configuration file.

Configuration

View docs
{
    "mcpServers": {
        "youtube": {
            "command": "mcp-youtube",
            "args": []
        }
    }
}

You use this MCP server as a bridge between the Youtube API and your AI assistants. It lets you fetch closed captions from a video and create helpful outputs like video summaries, all while keeping you in control of how data is accessed.

How to use

Connect Claude Desktop or another MCP client to the Youtube MCP Server to access video data from YouTube. You can request actions like downloading closed captions for a specific video and generating a concise video summary. When you issue a request, the MCP server runs locally and returns structured content that your AI assistant can present to you.

How to install

Prerequisites: you need the uv tool installed on your system.

Install the Youtube MCP server with the following command:

uv tool install git+https://github.com/sparfenyuk/mcp-youtube

Additional sections

Configuration, security, and troubleshooting details are provided to help you set up and run the server smoothly.

Configure Claude Desktop to recognize the Youtube MCP server by editing the Clara Desktop configuration. Add a server entry that points to the mcp-youtube command so Claude Desktop can start the MCP pipeline.

Troubleshooting

If you see a connection issue, ensure the MCP server command is available in your PATH and that the configuration file is correct. You can try running the server directly to verify it starts without errors.