openocr-skills

Extract text, formulas, and tables from images and PDFs using OpenOCR.

Updated May 7, 2026

Search Tags:#ocr #pdf #images #text-extraction #layout-analysis #document-parsing #openocr

Ready to run

npx skills add https://github.com/HvH1110/OpenOCR --skill openocr-skills-hvh1110

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: openocr-skills
Source: https://github.com/HvH1110/OpenOCR/tree/main/docs
Command: npx skills add https://github.com/HvH1110/OpenOCR --skill openocr-skills-hvh1110

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

OpenOCR enables automated extraction of text, formulas, and tables from images, scanned documents, and PDFs, turning unstructured visuals into searchable, editable content.

Core Features & Use Cases

Text detection, recognition, end-to-end OCR, and universal recognition for mixed content (text, formulas, tables).
Document parsing with layout analysis to preserve structure, enabling archiving, data extraction, and digital workflows.
Use Case: Convert a batch of scanned invoices into structured data with a single command.

Quick Start

Provide an input image or PDF and specify the task (det, rec, ocr, unirec, or doc) to immediately extract text, formulas, tables, or the full document layout.

Frequently Asked Questions about openocr-skills

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I extract text and tables from scanned PDFs and images?▼

To extract text and tables from scanned PDFs and images, you provide the input file and specify a task mode like OCR or doc. The layout analysis preserves document structure, yielding searchable and editable content.

Can I run OCR on CPU or do I need a GPU for document parsing?▼

You can run OCR on either CPU or GPU for document parsing. The workflow supports ONNX for CPU deployment and PyTorch for GPU acceleration, allowing flexible batch OCR pipelines depending on your environment.

How do I batch process invoices to extract structured data?▼

Batch processing invoices to extract structured data is supported through automated OCR pipelines. You can convert a batch of scanned invoices into structured text, formulas, and tables with a single command.

Does layout-aware OCR recognize formulas and tables in scientific papers?▼

Layout-aware OCR recognizes formulas and tables in scientific papers using universal recognition mode. This handles mixed content extraction, accurately parsing complex document structures across multiple languages.

What is the difference between det, rec, ocr, and unirec modes for text extraction?▼

Det performs text detection, rec handles text recognition, and OCR executes end-to-end extraction. Unirec provides universal recognition for mixed content like formulas and tables, while doc mode parses full document layouts.