kordoc

Overview of kordoc

What it is

kordoc is a CLI and MCP server tool that converts Korean document formats (HWP 3.x, 5.x, HWPX, HWPML, PDF, XLS, XLSX, DOCX, PPTX, PNG, JPG, WebP) to Markdown and structured data. It can be used as a library, CLI, or MCP server. The project is open source under the MIT license.

Key features

• Document reading: converts documents to Markdown, IR, per-page text, and RAG chunks; includes local OCR
• Comparison and editing: block-level and cell-level diff, patch preserving HWPX/HWP formatting
• Writing: Markdown to HWPX conversion with official document presets, tables, equations, and charts
• Form filling: fills fields and input boxes, includes two standard draft document templates and stamping
• Review: preview and extract regions from HWPX/HWP, structure and notation checking, personal information masking

Use cases

• Converting Korean government or office documents (HWP, HWPX, PDF, Office) to Markdown for AI processing or archiving
• Automating form filling in official draft documents
• Generating HWPX documents from Markdown with preset templates
• Comparing document versions at block or cell level
• Building RAG pipelines with chunked output from Korean documents

Getting started

Node.js 20 or later is required. Install via npm:bashnpm install kordocCLI can be run without installation using npx kordoc. To connect to AI agents, run:bashnpx -y kordoc setup

License and pricing

License: MIT. The project is open source.

Based on the project's public README and repository metadata.