Skip to main content
MCP server CX Codex

youtube-video-analyzer — Claude Code MCP server: YouTube analysis

youtube-video-analyzer — an MCP server that analyzes public YouTube videos via Google Gemini. Extracts speech, metadata, frames, handles long streams.

MIT ★ 5 updated 2026-05-21

Install into Claude Code · Platform CLI All methods and platforms ↓
claude mcp add --env GEMINI_API_KEY=<GEMINI_API_KEY> --env GEMINI_MODEL=<GEMINI_MODEL> --env YT_DLP_PATH=<YT_DLP_PATH> --transport stdio youtube-video-analyzer -- npx -y @ludylops/youtube-video-analyzer-mcp
Open on GitHub

What youtube-video-analyzer does

The server connects Gemini tools to Claude Code, Cursor, and other MCP clients. It can summarize videos, transcribe speech, fetch metadata through the YouTube Data API, and extract high-resolution frames using yt-dlp and ffmpeg. For long videos and VODs, the server offers two paths: MCP tasks (if the client supports them) or background jobs with status polling. When local tools are unavailable, the server switches to url_chunks strategy — no media downloads, but more Gemini calls.

Requires Node.js 20 or newer and a Gemini API key. YouTube Data API is only needed for metadata. Without yt-dlp and ffmpeg in PATH, the server cannot extract frames or process long videos via local files — only the URL strategy remains.

The server sends video URLs, prompts, and media fragments to Gemini. Private, deleted, age-restricted, and region-blocked videos may fail.

What it helps with

  • summarize a long stream or podcast
  • transcribe speech from a video
  • extract a frame from a video at a timestamp
  • fetch video metadata via YouTube Data API
  • ask a follow-up question about an already analyzed video

Example prompts for the agent

  • Analyze this video and extract key points
  • Extract a frame at 2:30 from this video
  • Fetch video metadata and show the duration

How to install youtube-video-analyzer

Platform CLI

claude mcp add --env GEMINI_API_KEY=<GEMINI_API_KEY> --env GEMINI_MODEL=<GEMINI_MODEL> --env YT_DLP_PATH=<YT_DLP_PATH> --transport stdio youtube-video-analyzer -- npx -y @ludylops/youtube-video-analyzer-mcp

Replace <…> with your own values

Step by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more

Questions about youtube-video-analyzer

How to install youtube-video-analyzer in Claude Code?

Run claude mcp add youtube-video-analyzer -e GEMINI_API_KEY=<GEMINI_API_KEY> -- npx -y @ludylops/youtube-video-analyzer-mcp in your terminal. Gemini key is required, YouTube API is only needed for metadata.

What does youtube-video-analyzer do?

Connects Google Gemini to MCP clients for analyzing public YouTube videos: summarizes content, transcribes speech, fetches metadata and frames, handles long videos via tasks or background jobs.

Can I analyze long streams without yt-dlp?

Yes, via url_chunks strategy. It skips local media downloads but requires more Gemini calls. For local file processing, you need yt-dlp and ffmpeg in PATH.

Similar MCP servers

All MCP servers for AI and ML →

Project sponsors

Companies that keep this analytics open