Webcam/Screenshot Capture MCP server

Enables capturing and analyzing live webcam images and screenshots for real-time visual context in AI applications.
Back to servers
Provider
evalstate
Release date
Jan 23, 2025
Language
TypeScript
Package
Stats
243 downloads
22 stars

The MCP Webcam Server allows you to connect your webcam to Claude Desktop or other MCP clients, enabling Claude to see through your camera and take screenshots. It provides tools for capturing frames from your webcam and taking screenshots, making it possible for Claude to visually interact with your environment.

Installation

To install the MCP Webcam Server:

  1. Make sure you have a recent version of NodeJS installed on your system

  2. Add the following configuration to the mcpServers section of your claude_desktop_config.json file:

"webcam": {
  "command": "npx",
  "args": [
    "-y",
    "@llmindset/mcp-webcam"
  ]
}

This setup works on both Windows and MacOS with Claude Desktop version 0.78 or later.

Port Configuration

By default, the server runs on port 3333 to avoid conflicts with other services. You can specify a different port as an argument if needed.

Usage

Basic Usage

  1. Start Claude Desktop
  2. Connect to http://localhost:3333 in your browser
  3. You can now ask Claude to interact with your webcam using natural language prompts:
    • "Get the latest picture from my webcam"
    • "Claude, take a look at what I'm holding"
    • "What color top am I wearing?"

Features

  • Live Capture: Claude can access the current view from your webcam
  • Freeze Frame: You can "freeze" the current image, which will be sent to Claude instead of a live capture
  • Screenshots: Ask Claude to take screenshots of your screen
    • Navigate to the desired browser window before the capture
    • Screenshots are automatically resized to be manageable for Claude (especially useful for 4K screens)
    • Note that screenshot functionality does not work on Safari as it requires human initiation

Using the "What am I holding?" Button

This button sends a sampling request to the client containing the webcam image and the question "What is the User holding?"

Note: Claude Desktop does not currently support Sampling. If you need a client that can handle multi-modal sampling requests, you can try https://github.com/evalstate/fast-agent/

Additional Resources

For more information about handling files and resources in LLM/MCP Chat Applications, see the article at https://llmindset.co.uk/posts/2025/01/mcp-files-resources-part1/

How to add this MCP server to Cursor

There are two ways to add an MCP server to Cursor. The most common way is to add the server globally in the ~/.cursor/mcp.json file so that it is available in all of your projects.

If you only need the server in a single project, you can add it to the project instead by creating or adding it to the .cursor/mcp.json file.

Adding an MCP server to Cursor globally

To add a global MCP server go to Cursor Settings > MCP and click "Add new global MCP server".

When you click that button the ~/.cursor/mcp.json file will be opened and you can add your server like this:

{
    "mcpServers": {
        "cursor-rules-mcp": {
            "command": "npx",
            "args": [
                "-y",
                "cursor-rules-mcp"
            ]
        }
    }
}

Adding an MCP server to a project

To add an MCP server to a project you can create a new .cursor/mcp.json file or add it to the existing one. This will look exactly the same as the global MCP server example above.

How to use the MCP server

Once the server is installed, you might need to head back to Settings > MCP and click the refresh button.

The Cursor agent will then be able to see the available tools the added MCP server has available and will call them when it needs to.

You can also explictly ask the agent to use the tool by mentioning the tool name and describing what the function does.

Want to 10x your AI skills?

Get a free account and learn to code + market your apps using AI (with or without vibes!).

Nah, maybe later