This MCP screenshot server, developed by kazuph, provides AI assistants with the ability to capture and analyze screen content on macOS. Built with TypeScript and leveraging libraries like Tesseract.js for OCR, it offers capabilities for taking screenshots of specific screen regions (left, right, or full) and performing text recognition on the captured images. The server integrates with the macOS system through AppleScript, automatically saving screenshots to dated folders in the user's Downloads directory. It's particularly useful for AI assistants focused on visual analysis, enabling use cases such as automated UI testing, content extraction from applications, and visual data processing in macOS environments.
Takes a screenshot and performs OCR. Options: region (string: 'left'/'right'/'full', default: 'left'), format (string: 'json'/'markdown'/'vertical'/'horizontal', default: 'markdown')
No reviews yet. Be the first to review!
Sign in to join the conversation
Highlight AIStart the server with node to access it via any client or IDE.
node path/to/downloaded/file.mjs