SnapStack MCP server

SnapStack is a small helper that runs on your own computer. It collects screenshots you take of your browser tabs and lets your AI assistant look at them when you ask. It is handy if you often want your AI to read something on a web page without copying and pasting the text.

What this MCP server does

You take a screenshot of a browser tab using a companion browser extension. The extension sends the picture to SnapStack, which runs quietly on your computer and saves it in a folder. When you ask your AI about your screenshots, your AI asks SnapStack for the list. SnapStack gives back a list of what is available, with the file locations, and your AI reads only the ones it needs. Nothing is sent to the internet; everything stays on your machine.

Click to zoom

What you can do with it

• Ask your AI to list the screenshots you have waiting
• Have your AI read what is on a captured page
• Ask your AI to summarize a page you screenshotted
• Check how many screenshots are waiting without opening the folder
• Delete specific screenshots when you are done with them
• Clear the whole stack of screenshots in one go

Try asking your AI

• “List my pending screenshots and tell me what each one shows.”
• “Read screenshot number 2 and summarize the article.”
• “How many screenshots do I have waiting?”
• “Delete screenshots 1 and 3, then clear the rest.”

What it gives back to you

It gives back a list of your screenshots with a number, the file location, the size, and some details like the page address and title. It does not send the actual image inside the chat; your AI opens the files it needs from your computer. When you ask it to delete, it tells you what was removed. It also answers simple questions like how many screenshots are waiting.

Before you start

Who it's for

Anyone who wants their AI to read web pages they screenshot, like researchers, writers, support staff, or anyone who copies text from pages into chat.