claude-video-vision MCP Server

Use This MCP server To

Get a summary of a long YouTube video without watching it Find out what's said at a specific moment in a video Turn a recorded meeting or lecture into written notes Check what text or numbers appear on screen in a clip Ask what language or topic a tutorial video covers Pull the key moments out of a long recording

README

claude-video-vision

Claude Code Video Vision

Give Claude the ability to watch and understand videos.

A Claude Code plugin that extracts frames via ffmpeg and processes audio via multiple backends (Gemini API, local Whisper, or OpenAI API). Claude receives frames as images and audio transcription with timestamps — the plugin is a perception layer, not an interpretation layer.

Features

Multimodal perception — Claude sees video frames directly and reads audio transcriptions with timestamps
YouTube URL support — Pass a YouTube URL directly; the MCP server downloads it with yt-dlp, preserving source metadata and captions for context
Flexible backends — Choose between cloud APIs or fully local processing
Adaptive extraction — Claude adjusts fps, time range, and resolution based on your question
Auto-installation — Whisper models download automatically on first use
Interactive setup wizard — /setup-video-vision walks you through configuration

Quick Start

1. Install the plugin

Inside Claude Code, run these commands one at a time:

/plugin marketplace add https://github.com/jordanrendric/claude-video-vision

Then:

/plugin install claude-video-vision

The MCP server will auto-install via npx from npm on first use — no build step required.

Alternative: local development

git clone https://github.com/jordanrendric/claude-video-vision.git
claude --plugin-dir /path/to/claude-video-vision

2. Configure

Inside Claude Code, run the interactive wizard:

/setup-video-vision

It will walk you through backend selection, whisper configuration (if local), frame options, and dependency verification.

claude-video-vision FAQ

Can I use this to summarize a long YouTube video?

Yes — give it the video link and ask for a summary; it watches the frames and reads the audio, then tells you what happens and when.

Can I use this to find out what's said at a specific timestamp in a video?

Yes — just ask about that moment, for example "what's said at 1:30?", and it will look closely at that part of the video.

Which apps does this work in?

It's built specifically for Claude Code, Anthropic's coding assistant tool, rather than Claude Desktop, ChatGPT, or Gemini's chat apps.

Do I need an account or a key to use this?

It depends on your setup. You can grab a free key from Google to power the Gemini option, use a paid OpenAI key instead, or skip keys altogether by running everything locally on your own computer for free — all three options work inside Claude Code.

How hard is it to set up?

You install it with a couple of quick commands inside Claude Code, then a setup wizard walks you through picking how to process audio and checking that everything else you need is installed.

Does it work with videos I already have saved on my computer?

Yes — point it at any video file on your computer, or give it a YouTube link, and it works the same way either way.

Will it work with a long video, like a full lecture?

Yes — it automatically adjusts to scan through a full hour-long recording at a lighter level of detail so it doesn't take forever.

Do I need to install anything besides the plugin itself?

Yes — you'll need a free tool called ffmpeg for reading video frames, plus either a free Google key, an OpenAI key, or a free local speech tool, depending on which option you pick.