MCP server · Design tools
AI Vision MCP server
Lets your AI look at images and videos, describe them, compare them, and check design quality.
This is a small helper that gives your AI the ability to look at pictures and videos. You point it at a screenshot, a photo, or a YouTube link, ask a question, and it tells you what it sees. It is handy for anyone who reviews designs, checks screenshots, or needs a quick read on visual content.
What is an MCP server? The 30-second version
On its own, your AI can only chat with you using text. An MCP server is a small helper program that gives your AI a new skill or a connection to another service. This one connects your AI to Google's Gemini and Vertex AI vision models, so it can actually look at images and videos. When you ask about a picture, your AI passes the request to this helper, and the helper sends it to Google's vision service for you.
What this MCP server does
You give your AI an image or video, either as a web link, a file on your computer, or pasted data. Your AI sends that along with your question to this helper. The helper passes it to Google's Gemini or Vertex AI service, which looks at the picture or video and writes back an answer. Your AI then shows you that answer in the chat. For object detection, it can also draw boxes around the things it finds and save a marked-up copy of the image.
Click to zoomWhat you can do with it
- Describe what is in a photo or screenshot
- Compare two to four images and point out the differences
- Pull out colors, spacing, and fonts from a design screenshot
- Check a design for contrast and accessibility problems
- Detect objects in an image and save a version with boxes drawn on it
- Summarize what happens in a YouTube video or a local video file
- Analyze visual hierarchy and where the eye goes on a page
Try asking your AI
- “What is this image about? Describe what you see in detail.”
- “Compare these two screenshots and tell me what changed.”
- “Extract all the design tokens from this screenshot.”
- “Audit this design for accessibility and check WCAG AA contrast.”
What it gives back to you
You get back a written answer in the chat, like a description, a comparison, a list of colors, or a list of design issues. For object detection, you also get a list of the objects found with their positions, plus a saved image with boxes drawn around them. For design audits, you get colors, contrast checks, and suggestions for improvement. Everything shows up as normal text in your conversation.
Before you start
What you need
- Node.js installed on your computer (the helper runs through a tool called npx)
- A Google AI Studio API key (a kind of password for apps; you can get one free from the Google AI Studio website)
- An MCP client like Claude Desktop, Claude Code, Cursor, or Cline
- For the Vertex AI option instead: a Google Cloud account with a service account and a storage bucket
Good to know
Your images and videos are sent to Google's AI service for analysis, so do not use it with private or sensitive material you would not share with that service.
Install it with your AI
Add AI Vision MCP server to your AI, no technical skills needed
You don't install anything by hand. You copy one prompt, paste it into an AI that can work on your computer, and it checks, installs and connects the server for you, asking you when it needs something.
Sign in to get the install prompt
Members get a ready-made prompt that lets the Claude desktop app check AI Vision MCP server, install it and connect it for them, step by step. You don't need any technical skills: you copy, paste and answer a few questions. Your connected AI can also find and install any of the 4,066 MCP servers here for you.
Who it's for
Designers, QA testers, and anyone who reviews screenshots, mockups, or videos and wants a quick AI read on them.
