Which image formats are supported by this skill?

The image-ocr skill is specifically optimized to batch-process JPG and JPEG image files from a designated directory.

Can I process a specific range of files in a directory?

Yes, you can use the --start and --limit arguments to process a specific subset of images within a sorted directory list.

Does this work on Windows and Linux?

Yes, while the Vision and MLX backends are macOS-specific, the Tesseract backend provides cross-platform support for Windows and Linux users.

What is the benefit of the 'vision' backend?

The vision backend uses Apple's native framework, which is on-device, GPU-accelerated, and extremely fast for users on Apple Silicon hardware.

How do I improve OCR accuracy for low-quality images?

You can use the --preprocess flag to enable grayscale conversion, autocontrast, and upscaling before the OCR process begins.

Image OCR Toolkit

Name: Image OCR Toolkit
Author: varunr89

byvarunr89

0•

Productivity & Workflow

Extracts text from directories of JPG and JPEG images into structured Markdown files using multiple high-performance OCR backends.

The Image OCR skill for Claude Code is a robust utility designed for batch-processing image text extraction with high efficiency. It provides a flexible architecture supporting Apple's native Vision framework for GPU-accelerated performance on macOS, Tesseract for cross-platform reliability, and MLX OCR for Apple Silicon optimization. Perfect for digitizing whiteboards, screenshots, or scanned documents, this skill automates the conversion of visual data into searchable Markdown, featuring parallel processing, image preprocessing for better accuracy, and multi-language support to streamline your documentation and data collection workflows.

Key Features

01Automated image preprocessing including upscaling and autocontrast

02Unified Markdown output for easy integration into existing docs

03Multi-backend support including Apple Vision, Tesseract, and MLX OCR

04Batch-processing for entire directories of JPG/JPEG images

05Parallel processing with configurable workers for high-speed extraction

060 GitHub stars

Use Cases

01Extracting data from UI screenshots for technical specifications and audits

02Bulk digitizing scanned paper documents and archives into searchable text

03Converting whiteboard session photos or meeting screenshots into project documentation

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add varunr89/claude-marketplace image-ocr

For use in Claude.ai and ChatGPT

Key Features

01Automated image preprocessing including upscaling and autocontrast

02Unified Markdown output for easy integration into existing docs

03Multi-backend support including Apple Vision, Tesseract, and MLX OCR

04Batch-processing for entire directories of JPG/JPEG images

05Parallel processing with configurable workers for high-speed extraction

060 GitHub stars

Use Cases

01Extracting data from UI screenshots for technical specifications and audits

02Bulk digitizing scanned paper documents and archives into searchable text

03Converting whiteboard session photos or meeting screenshots into project documentation