OpenDocKit: Open-Source Document Converter Supporting HWP

·2026.11.13 17:28

Key point

OpenDocKit, a document converter supporting HWP, has been released as open source with integration for LangChain, LlamaIndex, and MCP.

Released OpenDocKit, a Python library that converts various formats such as PDF, Office, HWP/HWPX, and EPUB into Markdown for LLMs, under the MIT license
Directly implemented an HWP 5.x parser to support outlines, merged cells, footnotes, and Korean formulas; confirmed in comparative benchmarks as the only tool capable of reading HWP/HWPX
Passed 49 out of 49 structural inspection items and successfully extracted text from 667 out of 743 test documents
Supports LangChain and LlamaIndex loaders as well as an MCP server, enabling direct document access in agent environments
Lacks built-in OCR, so scanned PDFs or images require connection to a vision model API, and it is slower than Kreuzberg

This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.

Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.