kordoc

View on GitHub

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

★ 2.3K420 forksTypeScriptMITchrisryugj

TypeScript CLI + MCP server that parses Korean documents (HWP 3.x/5.x, HWPX, HWPML, PDF, XLS/X, DOCX, images) into Markdown, with lossless HWPX roundtrip, form filling, document diff, RAG chunking, OCR, and PII redaction. Registers 17 document tools for Claude Desktop, Cursor, and Codex.

Use Cases

Convert HWP/HWPX/PDF/Office files to Markdown for LLM ingestionRegister 17 document tools in Claude Desktop, Cursor, or Codex via MCPAuto-fill Korean government form templates while preserving original formattingGenerate HWPX government reports and drafts from MarkdownDiff two documents (including HWP vs HWPX) to build a revision comparison tableChunk documents with breadcrumb headings and real page numbers for RAG citationsOCR scanned PDFs and images locally without API keysMask resident IDs, phone numbers, cards, and accounts while keeping layoutRender HWP/HWPX layout to SVG/PNG/PDF and crop table regionsRun offline deployments for air-gapped document pipelinesRestore merged/nested tables from HWPX and PDF into structured MarkdownInstall as a Claude Code skill plugin for automatic .hwp/.hwpx handling

Built With

Language: TypeScript
Frameworks: @modelcontextprotocol/sdk · commander · jszip · markdown-it · zod · puppeteer-core · tsup · node.js · cfb · @xmldom/xmldom

kordoc

Use Cases

Built With

Tags