Macos Screenshot

🚀 MCP Screenshot

An MCP server for capturing screenshots and performing OCR text recognition.

🚀 Quick Start

This is an MCP server designed to capture screenshots and perform OCR text recognition. It offers a range of features and can be easily integrated into your workflow.

✨ Features

• Screenshot capture (left half, right half, full screen)
• OCR text recognition (supports Japanese and English)
• Multiple output formats (JSON, Markdown, vertical, horizontal)

📦 Installation

npx -y @kazuph/mcp-screenshot

📚 Documentation

Claude Desktop Configuration

Add the following configuration to your claude_desktop_config.json file:

{
  "mcpServers": {
    "screenshot": {
      "command": "npx",
      "args": ["-y", "@kazuph/mcp-screenshot"],
      "env": {
        "OCR_API_URL": "http://localhost:8000"  // Base URL of the yomitoku API
      }
    }
  }
}

Environment Variables

Tool Specification - capture

Capture a screenshot and perform OCR.

Options:

• region: Screenshot area ('left'/'right'/'full', default: 'left')
• format: Output format ('json'/'markdown'/'vertical'/'horizontal', default: 'markdown')

🔧 Technical Details

OCR Engines

This server uses two OCR engines:

1. yomitoku

   • Primary OCR engine
   • High-precision Japanese text recognition
   • Runs as an API server

2. Tesseract.js

   • Fallback OCR engine
   • Used when yomitoku is unavailable
   • Supports Japanese and English recognition

💻 Usage Examples

Basic Usage

You can use it by instructing Claude as follows:

Please capture the left half of the screen and recognize the text in it.

📄 License

MIT

Author

kazuph