Korean HWP to Markdown Converter
Converts Korean HWP, HWPX, and HWPML documents into LLM-ready Markdown or structured JSON while preserving document architecture.
This skill provides a specialized harness for processing Korean word processor files (Hangul) directly within Claude's environment. By leveraging the kordoc engine, it extracts text, headings, tables, and media from .hwp and .hwpx formats, transforming them into clean Markdown or JSON. It is particularly effective for preparing government or enterprise documents for RAG (Retrieval-Augmented Generation) and AI context, featuring automated table flattening and Hancom-specific character cleanup to ensure high-fidelity data extraction that is readable by large language models.
主要功能
01Full support for HWP 5.x binary, HWPX, and HWPML formats
021,082 GitHub stars
03Automated GFM table flattening for better Markdown compatibility
04Cleanup of Hancom Private Use Area (PUA) glyphs for standard rendering
05High-fidelity extraction of headings, tables, lists, and footnotes
06Vendor-agnostic agent harness designed for project-specific engineering standards
使用场景
01Extracting structured data from enterprise Hangul reports for database ingestion
02Preparing legacy .hwp files for Retrieval-Augmented Generation (RAG) systems
03Transforming Korean government documents into Markdown for AI analysis