macOS Screenshot
About
Captures and analyzes macOS screen content using TypeScript and OCR, enabling automated UI testing and visual data processing.
Details
- Author
- kazuph
- Repository
- kazuph/mcp-screenshot
- GitHub stars
- 21
- License
- MIT License
- Categories
- Productivity, Automation, Developer Tools, Other, Design, File Management, AI, Media, Frontend, Infrastructure
Jump to
- Screenshot capture (left half, right half, full screen)
- OCR text recognition (supports Japanese and English)
- Multiple output formats (JSON, Markdown, vertical, horizontal)
Setting up with Highlight
This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:
- Download and install Highlight from highlightai.com/download
- Navigate to the plugins tab and select "Add Custom Plugin"
-
Configure the plugin with the settings below
Plugin Name
macOS ScreenshotCommand (node, npx, python, etc.)npxArguments-
Argument 1
-y -
Argument 2
@kazuph/mcp-screenshot
Environment-
OCR_API_URL
http://localhost:8000
Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.
-
Argument 1
- Enable "Start Automatically" if you want the plugin to start when Highlight launches
From the repository
npx -y @kazuph/mcp-screenshot
Add the following configuration to your claude_desktop_config.json:
{
"mcpServers": {
"screenshot": {
"command": "npx",
"args": ["-y", "@kazuph/mcp-screenshot"],
"env": {
"OCR_API_URL": "http://localhost:8000" // yomitoku API base URL
}
}
}
}
| Variable Name | Description | Default Value |
|--------------|-------------|---------------|
| OCR_API_URL | yomitoku API base URL | http://localhost:8000 |
You can use it by instructing Claude like this:
Please take a screenshot of the left half of the screen and recognize the text in it.
capture
Takes a screenshot and performs OCR. Options: region (string - 'left'/'right'/'full', default: 'left'), format (string - 'json'/'markdown'/'vertical'/'horizontal', default: 'markdown')
Claude Desktop / Cursor
Paste into your MCP client config file to install this server.
{
"mcpServers": {
"macos screenshot": {
"env": {
"OCR_API_URL": "http://localhost:8000"
},
"args": [
"-y",
"@kazuph/mcp-screenshot"
],
"command": "npx"
}
}
}Linux
{
"env": {
"OCR_API_URL": "http://localhost:8000"
},
"args": [
"-y",
"@kazuph/mcp-screenshot"
],
"command": "npx"
}Macos
{
"env": {
"OCR_API_URL": "http://localhost:8000"
},
"args": [
"-y",
"@kazuph/mcp-screenshot"
],
"command": "npx"
}Windows
{
"env": {
"OCR_API_URL": "http://localhost:8000"
},
"args": [
"/c",
"npx",
"-y",
"@kazuph/mcp-screenshot"
],
"command": "cmd"
}An MCP server that captures screenshots and performs OCR text recognition.
- Screenshot capture (left half, right half, full screen)
- OCR text recognition (supports Japanese and English)
- Multiple output formats (JSON, Markdown, vertical, horizontal)
- Primary OCR engine
- High-accuracy Japanese text recognition
- Runs as an API server
- Fallback OCR engine
- Used when yomitoku is unavailable
- Supports both Japanese and English recognition
Add the following configuration to yourclaude_desktop_config.json:
{ "mcpServers": { "screenshot": { "command": "npx", "args": ["-y", "@kazuph/mcp-screenshot"], "env": { "OCR_API_URL": "http://localhost:8000" // yomitoku API base URL } } } }
You can use it by instructing Claude like this:
Please take a screenshot of the left half of the screen and recognize the text in it.
- region: Screenshot area ('left'/'right'/'full', default: 'left')
- format: Output format ('json'/'markdown'/'vertical'/'horizontal', default: 'markdown')
MCP server for the Computer-Use Agent (CUA), allowing you to run CUA through Claude Desktop or other MCP clients.
MCP server for Mobile Development and Automation | iOS, Android, Simulator, Emulator, and Real Devices
A simple, self-contained notes system with resources, tools, and prompts.
Control your computer's mouse, keyboard, and perform OCR using PyAutoGUI and RapidOCR. Works on Windows, with potential support for other platforms.
MCP allowing your Agent to request human validation on webapps and apps.
Automate desktop actions like mouse control, keyboard input, and taking screenshots.
Unified mobile E2E MCP — 28 tools (mobile-mcp + Maestro + RN-debugger parity)
Provides comprehensive browser automation for QA testing capabilities.
Capture and analyze screenshots using the Claude Vision API.
Paid remote MCP endpoint for synthetic user testing, UX validation status, analytics, checkout, pricing, and site-readiness reporting.
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
