youtube

This MCP server provides YouTube data access via the Model Context Protocol, exposing a toolset that lets language models query YouTube channels and their metadata. It supports two transport modes: stdio for local, self-contained use and Server-Sent Events (SSE) for remote access. The core capability is a tool named get_youtube_channel (used as an MCP tool) which can search channels by name and return detailed information such as channel statistics and metadata. To use with GitHub Copilot Chat, configure the MCP server in your Copilot Chat MCP settings, select the appropriate transport, and invoke the get_youtube_channel tool in natural language prompts like “Find the channel named X” or “Get details for channel Y.” The inspector utility can also help you experiment with the available tools and their input/output schemas before integration.

Prerequisites:

• Node.js v18.0.0 or higher
• npm (or yarn) installed
• YouTube Data API v3 key (YOUTUBE_API_KEY)
• Git (optional for cloning)

Installation steps:

Option A: MCP Server with stdio transport (mcp-stdio)

1. Clone the repository and open the stdio folder: git clone https://github.com/0GiS0/youtube-mcp-server.git cd youtube-mcp-server/mcp-stdio

2. Install dependencies: npm install

3. Configure environment: cp .env-sample .env Edit .env and add your YouTube API key: YOUTUBE_API_KEY=your_api_key_here

4. Build the project (if required by the repo setup): npm run build

Option B: MCP Server with SSE transport (mcp-remote)

1. Clone the repository and open the remote folder: git clone https://github.com/0GiS0/youtube-mcp-server.git cd youtube-mcp-server/mcp-remote

2. Install dependencies: npm install

3. Configure environment: cp .env-sample .env Edit .env and add your YouTube API key: YOUTUBE_API_KEY=your_api_key_here

4. Build and run the server: npm run build npm start

Both setups expose the MCP tools and can be connected to via GitHub Copilot Chat by adding them to your MCP configuration.

Tips and common considerations:

• Ensure your YOUTUBE_API_KEY has the YouTube Data API v3 enabled in the Google Cloud Console.
• The stdio transport is best for local testing and quick experiments, while SSE is suited for remote integration scenarios.
• If you encounter rate limiting from the YouTube API, consider implementing caching or rate-limiting logic at the client side and verify API quotas in Google Cloud Console.
• Use the MCP Inspector utility (npx @modelcontextprotocol/inspector) to explore the available tools and their input/output schemas before integrating with Copilot Chat.
• When using Copilot Chat, provide clear prompts that specify the tool name (e.g., get_youtube_channel) and required parameters to improve reliability.