mcp-youtube-transcript

MCP.Pizza Chef: jkawamoto

Paste a video link and this add-on fetches the spoken words, with or without timestamps, plus the video's title and the languages its captions come in. Long videos are split into parts so nothing gets cut off. It works in Claude Desktop, Cursor, LM Studio, and Goose, needs no account or sign-in, and can route through a proxy service if YouTube blocks your connection. Claude Desktop users can drag in a ready-made bundle.

Notes
Web/Research
Writing

Use This MCP server To

Summarize an hour-long talk without watching it Pull quotes with timestamps from a video I need to cite Check whether a video covers my topic before watching Read a lecture's captions in another language Turn a tutorial into written step-by-step notes

README

YouTube Transcript MCP Server

uv Python Application pre-commit GitHub License Dockerhub

This MCP server retrieves transcripts for given YouTube video URLs.

Tools

This MCP server provides the following tools:

get_transcript

Fetches the transcript of a specified YouTube video.

Parameters
  • url (string): The full URL of the YouTube video. This field is required.
  • lang (string, optional): The desired language for the transcript. Defaults to en if not specified.
  • next_cursor (string, optional): Cursor to retrieve the next page of the transcript.

get_timed_transcript

Fetches the transcript of a specified YouTube video with timestamps.

mcp-youtube-transcript FAQ

Is this still maintained?
Yes — the project had code changes in August 2026 and publishes regular releases.
Which apps does it work in?
Claude Desktop, Cursor, LM Studio, and Goose, plus anything else that reads a standard tool-server config.
Do I need a token or a YouTube account?
No — no sign-in at all, though it can use a paid proxy service if YouTube blocks you.
How hard is setup?
Claude Desktop and LM Studio have one-click installs; other apps need a short config snippet pasted in.
Can I use this to summarize a long video?
Yes — it splits very long transcripts into parts so the whole thing gets read.
What if a video has no captions?
There is nothing to fetch — it reads existing captions rather than listening to the audio.
Can I get the transcript in another language?
Yes — ask for a language, and it can also list which languages a video offers.