YouTube Insights

Provides YouTube data access via MCP, including transcripts, keyword-based video discovery, and channel information.

javascript

3

GitHub Stars

javascript

Language

4 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "dabidstudio-youtubeinsights-mcp-server": {
      "command": "uvx",
      "args": [
        "youtubeinsights-mcp-server"
      ],
      "env": {
        "YOUTUBE_API_KEY": "your-api-key"
      }
    }
  }
}

You can run a YouTube Insights MCP Server to extract transcripts from videos, discover videos by keyword, and fetch channel information, then integrate these capabilities into your agent workflows with FastMCP. This server simplifies accessing YouTube data through a standardized MCP interface so your agents can perform transcript extraction, video search, and channel info retrieval efficiently.

How to use

Use this MCP server by configuring it in your MCP client to run locally or via an MCP runner. You can request a video transcript with subtitles, search for videos by keywords and obtain metadata such as views and thumbnails, and retrieve channel information based on any YouTube video URL. Once the MCP server is running, your client can call the available tools to perform these actions and integrate the results into your agent workflows.

How to install

Prerequisites you need installed on your system before running the MCP server are Node.js and a runtime capable of executing MCP servers (such as uvx or npx-based runners). Ensure you have an active YouTube Data API key to access transcripts and video data.

// Option 1: Run via uvx (recommended for local development and quick testing)
{
  "mcpServers": {
    "youtubeinsights": {
      "command": "uvx",
      "args": ["youtubeinsights-mcp-server"],
      "env": {
        "YOUTUBE_API_KEY": "your-api-key"
      }
    }
  }
}

// Option 2: Run a development flow using uv with a directory and explicit run command
{
  "mcpServers": {
    "youtubeinsights": {
      "command": "uv",
      "args": [
        "--directory",
        "path/to/youtubeinsights-mcp-server",
        "run",
        "youtubeinsights-mcp-server"
      ],
      "env": {
        "YOUTUBE_API_KEY": "your-api-key"
      }
    }
  }
}

Configuration

The server supports two local runtime configurations. Choose one to start the MCP server locally and connect it to your MCP client.

Common environment variable you need is the YouTube Data API key. Provide it in the environment block when starting the server.

Troubleshooting

If you cannot reach the MCP server from your client, verify that the runtime process is running and that the YOUTUBE_API_KEY is set correctly. Check the console for any API quota or authentication errors from the YouTube Data API.

Notes

Two local execution methods are shown. You can choose either depending on your development workflow. The APIs exposed by the server cover transcripts extraction for multiple languages, keyword-based video discovery with metadata, and channel information retrieval.

Available tools

get_youtube_transcript

Extract full transcripts (subtitles) from a YouTube video URL, supporting multiple languages.

search_youtube_videos

Search for videos on YouTube by keyword and return metadata such as views, likes, and thumbnails.

get_channel_info

Retrieve channel information and recent uploads based on a YouTube video URL.