YouTube MCP(Model Context Protcol) Server
@PrajwalPrashanth
MCP server to fetch and extract transcripts from YouTube videos. Enable AI/LLMs to have access to transcripts for further actions like summarize, write blog..
Overview
What is YouTube MCP Server?
YouTube MCP Server is a Model Context Protocol (MCP) server that provides a tool to fetch and extract transcripts from YouTube videos. It enables AI language models to access complete text transcripts from YouTube URLs, process video content through transcripts, and reference video information in conversations.
How to use YouTube MCP Server?
Install the dependencies and add the server to Claude Desktop using uv run mcp install -e . server.py -n "youtube-mcp". Once installed, a tool icon appears in the Claude Desktop UI; provide a YouTube video URL and ask for actions such as summarization using the transcript.
Key features of YouTube MCP Server
- Fetches and extracts transcripts from YouTube video URLs
- Provides complete text transcripts to AI language models
- Acts as a bridge between AI models and YouTube content
- Automatically formats extracted transcript text
Use cases of YouTube MCP Server
- Summarize the content of a YouTube video using its transcript
- Analyze or discuss specific parts of a video’s spoken content
- Extract information from educational or tutorial videos for reference
FAQ from YouTube MCP Server
What exactly does this server do?
It fetches the transcript text of a YouTube video and provides it to an AI model, allowing the model to process and discuss the video content.
What dependencies are required?
You need Python, the uv package manager, and the dependencies listed in pyproject.toml. The README provides install commands for macOS, Linux, and Windows.
How do I add this server to Claude Desktop?
Run uv run mcp install -e . server.py -n "youtube-mcp" from the cloned repository directory after setting up the virtual environment and installing dependencies.
How do I use it after installation?
Once the tool icon appears in Claude Desktop, simply provide a YouTube URL and ask the model to perform an action like summarizing the video’s transcript.
Is there any authentication or transport configuration required?
The README does not mention any authentication, API keys, or transport settings. Setup follows the standard MCP server configuration for Claude Desktop.
