kordoc

Why we included this project

Anyone who has dealt with Korean government or corporate paperwork knows how closed and awkward HWP files can be. Kordoc converts HWP, HWPX, PDF, and Office files into clean Markdown, which makes the content usable by LLMs for analysis, search, and automated workflows. Beyond plain extraction, it restores complex merged and nested tables without data loss, and adds practical extras like automatic form filling and a before-and-after diff for revised documents. It ships as both a command-line utility and an MCP server, so it plugs into AI clients like Claude Desktop, Cursor, or Claude Code for on-demand parsing. Teams handling official Korean documents, legal amendments, or office document pipelines get a useful bridge between legacy formats and modern AI tooling.