Why is GLM-OCR used specifically?

GLM-OCR is a specialized model optimized for high-accuracy text recognition from images, making it superior to general-purpose vision models for document digitizing.

How does it handle large PDF files?

For documents longer than 10 pages, the skill automatically runs the OCR process in the background so you can continue using Claude Code while it processes.

What dependencies are required to run pdf-ocr?

You need Ollama (with the glm-ocr model pulled), Poppler (for pdftoppm and pdfinfo utilities), and Python 3 installed on your system.

Does this skill send my PDF data to the cloud?

No, this skill runs entirely locally using Ollama and GLM-OCR, ensuring your documents remain private and never leave your machine.

Can I use this skill on Windows or Linux?

The skill is optimized for macOS (using sips), but it can function on Linux and Windows provided Ollama, Poppler, and Python are correctly configured in your environment.

Local PDF OCR to Markdown

Name: Local PDF OCR to Markdown
Author: AlteredCraft

byAlteredCraft

•

Content Management

Converts PDF documents into clean Markdown files using local GLM-OCR via Ollama for privacy-focused text extraction.

This skill provides a privacy-first, local-only solution for converting PDF files into structured Markdown. By leveraging GLM-OCR running on Ollama, it renders PDF pages into images and performs high-accuracy text recognition without sending sensitive data to external APIs. The skill automates dependency checks for Poppler and Ollama, handles background processing for long documents, and produces formatted Markdown output with page markers, making it an essential tool for developers and researchers digitizing documents directly within their terminal.

Key Features

01Automatic conversion of PDFs to structured Markdown output

02Optimized image rendering for maximum OCR character accuracy

03Automated dependency verification for Poppler and system tools

04Background execution support for multi-page document processing

05Privacy-first local processing via Ollama and GLM-OCR

0610 GitHub stars

Use Cases

01Digitizing legacy PDF documentation into searchable and editable Markdown files

02Extracting text from scanned documents for use in local RAG or AI knowledge bases

03Converting academic papers and technical reports into clean text formats locally

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add alteredcraft/claude-code-plugins pdf-ocr

For use in Claude.ai and ChatGPT

Download Skill