MCP.so
Sign In
Y

Youtube

@cactusaurelius

Repository

About Youtube

A Model Context Protocol (MCP) server for YouTube video transcripts and search.

#youtube-transcript#mcp-server#model-context-protocol#youtube-api#video-transcript-extraction#ai-tools#typescript#fastmcp#youtube-search#channel-analysis#video-content-analysis#transcript-chunking#claude-desktop#cursor-ide#ai-development-tools#youtube-data-extraction#video-processing#content-analysis#automated-transcription

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "youtube": {
      "command": "npx",
      "args": [
        "-y",
        "@inlustris/youtube-mcp@latest"
      ]
    }
  }
}

Tools

5

Retrieves the full transcript of a specified YouTube video. This tool is useful for understanding video content without watching it, or for extracting textual information from videos. FORMATTING GUIDANCE (optional - user instructions override): When creating summaries, consider using: **Key Points with Timestamps:** Use [MM:SS] or [HH:MM:SS] inline references. **Structure:** Break into logical sections. **Context:** Include video title and channel. Example: 'The speaker explains TypeScript generics [05:30] and shows practical examples [08:15].' This formatting is optional - always follow any specific user instructions instead.

Searches YouTube for videos matching the specified query. Returns a list of video results with title, video ID, and channel information. Results are sorted by rating by default for better quality content.

Searches YouTube for channels matching the specified query. You can specify a sort order for the results (default: rating).

Retrieves a list of videos from a specified YouTube channel. This tool is useful for getting all videos uploaded by a specific channel.

Captures a screenshot from a YouTube video at a specified timestamp. Requires yt-dlp for full-quality captures; falls back to lower-resolution YouTube storyboards (320x180 max) if yt-dlp is not installed.

Overview

What is Youtube?

A Model Context Protocol (MCP) server that provides tools for extracting YouTube video transcripts with timestamps and searching YouTube videos and channels. It is designed for AI tools and developers who need programmatic access to YouTube content.

How to use Youtube?

Install globally via npm: npm install -g @inlustris/youtube-mcp. Then add the server to your MCP client configuration (e.g., in .cursor/mcp.json) using the command npx -y @inlustris/youtube-mcp@latest. The server exposes tools that can be called from any MCP-compatible client.

Key features of Youtube?

  • Get video transcripts with timestamps
  • Search YouTube videos by query
  • Search YouTube channels by query
  • Get video lists from a specific channel
  • Chunking support for large transcripts

Use cases of Youtube?

  • Extract and analyze transcripts from educational or tutorial videos.
  • Search for relevant videos based on keywords and sort by criteria.
  • Discover YouTube channels matching a specific topic or interest.
  • List all recent videos from a known channel for content curation.
  • Process long transcripts by splitting them into manageable chunks.

FAQ from Youtube

What tools does the server provide?

It provides four tools: get_transcript, search_videos, search_channels, and get_channel_videos.

How do I install the server?

Install globally with npm: npm install -g @inlustris/youtube-mcp. Alternatively, run directly with npx -y @inlustris/youtube-mcp@latest.

What parameters are required for get_transcript?

The only required parameter is videoUrl (a YouTube video URL). Optional parameters include chunkSize, chunkBySilence, and silenceThreshold.

What sorting options are available for video search?

search_videos supports sorting by relevance, date, rating, viewCount, or title.

Is there a limit on results when getting channel videos?

Yes, the optional maxResults parameter for get_channel_videos can be set up to 200.

Frequently asked questions

What tools does the server provide?

It provides four tools: `get_transcript`, `search_videos`, `search_channels`, and `get_channel_videos`.

How do I install the server?

Install globally with npm: `npm install -g @inlustris/youtube-mcp`. Alternatively, run directly with `npx -y @inlustris/youtube-mcp@latest`.

What parameters are required for `get_transcript`?

The only required parameter is `videoUrl` (a YouTube video URL). Optional parameters include `chunkSize`, `chunkBySilence`, and `silenceThreshold`.

What sorting options are available for video search?

`search_videos` supports sorting by `relevance`, `date`, `rating`, `viewCount`, or `title`.

Is there a limit on results when getting channel videos?

Yes, the optional `maxResults` parameter for `get_channel_videos` can be set up to 200.

Comments

Basic information

Category

Media & Design

Transports

stdio

Author

cactusaurelius

Submitted by

Adam R

More Media & Design MCP servers

Instagram + TikTok Downloader & Transcriber

IYIA

Instagram + TikTok content tools inside your AI: download HD posts/reels/carousels, view Instagram stories anonymously, AI-transcribe videos (100+ languages, SRT/VTT/TXT). Requires a Repost PRO subscription.

SceneF — San Francisco Movie Showtimes

SceneF

Built by movie lovers, for movie lovers and their assistants. San Francisco is one of the last great moviegoing cities — single-screen neighborhood houses from the 1920s, repertory calendars that change nightly, 35mm and

Spotify MCP

varunneal

MCP to connect your LLM with Spotify.

606

Shotstack

Shotstack's official MCP server gives AI assistants direct access to the Shotstack video editing API, so you can produce video from a conversation instead of writing Edit JSON by hand. Available tools cover rendering vid

Framelink Figma MCP Server

GLips

MCP server to provide Figma layout information to AI coding agents like Cursor

15.3K