chrisryugj/kordoc

모두 파싱해버리겠다 — HWP·HWPX·PDF·Office 문서를 Markdown으로. 양식 자동 채우기와 신구대조를 갖춘 CLI·MCP 서버 | Convert Korean documents (HWP, HWPX, PDF, Office) to Markdown — CLI and MCP server with form filling and diff

What it solves

Kordoc is a high-precision document parsing library and toolset designed to handle the complex formatting of Korean government documents (HWP, HWPX, PDF, etc.) and convert them into structured data or Markdown. It solves the problem of losing structural integrity—especially complex tables—when converting legacy Korean word processor formats to AI-ready formats like Markdown for RAG (Retrieval-Augmented Generation) pipelines.

How it works

Kordoc provides a multi-format parser that supports HWP 3.x/5.x, HWPX, HWPML, PDF, and common office formats (XLS, XLSX, DOCX, PPTX) and images (PNG, JPG, WebP). It can be used as a Node.js library, a Command Line Interface (CLI), or an MCP (Model Context Protocol) server, allowing AI agents like Claude or Cursor to directly parse and compare documents.

Who it’s for

  • AI Engineers building RAG systems that need to ingest Korean government documents with high structural accuracy.
  • Developers needing to automate the conversion of HWP/HWPX files to Markdown or JSON.
  • Administrative Staff looking to automate the document generation and filling of official Korean government forms.

Highlights

  • Extreme Table Accuracy: Claims 100% structural match for tables in HWPX documents based on internal benchmarks.
  • AI Agent Integration: Built-in MCP server support for seamless integration with AI clients like Claude Desktop and Cursor.
  • RAG-Optimized: Includes specific output formats (like GFM pipe tables) for better indexing in RAG pipelines.
  • Bi-directional Conversion: Supports not only parsing to Markdown, but also generating HWPX files from Markdown using official document presets.
  • High-Performance PDF Parsing: Ranked 1st in a public PDF-to-Markdown benchmark (opendataloader-bench) with a score of 0.960.

Related

  • Project
    vivekVells/mcp-pandoc

    An MCP server that enables AI models to convert documents between multiple formats using Pandoc, supporting everything from Markdown to PDF and DOCX.

  • Project
    magicrew/doc7

    A document conversion tool that turns PDFs, Office files, and images into AI-ready Markdown using multimodal vision models, supporting local inference for privacy and cost-efficiency.

  • Project
    docling-project/docling

    Docling is an open‑source Python library/CLI that parses many document types (PDF, Word, slides, spreadsheets, HTML, e‑books, audio, video, email, etc.) into a unified, structured representation. It extracts layout, tables, code, formulas, images, and speech‑to‑text, then exports to Markdown, HTML, JSON, DocLang, DocTags, etc. Designed for generative‑AI workflows, it offers plug‑and‑play integrations with LangChain, LlamaIndex, Crew AI, Haystack, and supports local execution for privacy‑sensitive data.

  • Project
    Gostyan/docx-skill-4-cn-paper

    A toolset for AI Agents to create and edit Word documents following Chinese academic formatting standards, featuring Markdown-to-docx conversion and LaTeX-to-native-formula support.

  • Project
    SylphxAI/anymd

    anymd is a Rust‑based local server/CLI that converts PDFs, Word docs, slides, spreadsheets, web pages, images, audio, and video into clean, token‑efficient Markdown for AI agents. It runs fast (212× faster than docling on benchmarks), preserves layout (tables, headings, citations), works offline (no API keys), integrates via the Model‑Context‑Protocol, and offers optional Pro features for video timelines and citation checking.