kordoc: Instructions file for Codex

AGENTS.md

kordoc AGENTS.md is an instructions file for Codex, OpenCode from chrisryugj/kordoc. It costs 3,159 tokens per session, scanned A, original, MIT.

Repository instructions for kordoc, a tool that converts Korean HWP, HWPX, PDF, and Office documents into Markdown. Markdown is plain text with simple formatting markers.

In plain words
What is it for?
It guides development and release work on document parsing, form filling, document comparison, and PDF or Korean document quality checks.
Why use it?
It makes project-specific rules explicit for changes, testing, publishing, and document-quality checks. It also prevents unrelated files or work from being changed accidentally.

Instructions file for CodexOpenCode

Written for Codex and OpenCode: the file is AGENTS.md. Also seen: reads .claude/ paths; mentions CLAUDE.md; mentions Claude Code.

This is chrisryugj/kordoc's own configuration. It tells Codex and OpenCode how to work on kordoc itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything kordoc configures →

About the project

kordoc is a command-line tool and MCP server that reads Korean document formats and converts, compares, analyzes, fills, or generates them, including HWP, PDF, Office files, and images. Government workers and other users handling Korean documents use it to make their contents available as Markdown and automate document processing, while the catalogue entries provide integrations and workflows for coding agents.

chrisryugj/kordoc · 2,928 stars · on GitHub · pubax.org

Reuse

Borrowing it

Nothing to install: this file belongs to chrisryugj/kordoc. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/chrisryugj/kordoc/main/AGENTS.md
Clone the repo
git clone --depth 1 https://github.com/chrisryugj/kordoc

Made for: Codex, OpenCode.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for kordoc AGENTS.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/chrisryugj/kordoc/agents-md/github.svg)](https://agentmods.dev/instructions/chrisryugj/kordoc/agents-md)
Your own site
<a href="https://agentmods.dev/instructions/chrisryugj/kordoc/agents-md"><img src="https://agentmods.dev/badge/instructions/chrisryugj/kordoc/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for kordoc AGENTS.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/chrisryugj/kordoc/agents-md"><img src="https://agentmods.dev/badge/instructions/chrisryugj/kordoc/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 3,159 This file is loaded in full into every session.
When invoked 3,159 The same file — it is already loaded in full.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.03159 $0.03159
Opus 5.5 $0.01264 $0.01264
Sonnet 5.5 $0.00632 $0.00632
Haiku 4.5 $0.00316 $0.00316

Measured 6d ago against content hash d866dff212ce, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

kordoc AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

AGENTS.md · 97 lines

How it starts

The opening of the file, as written. The whole thing — 97 lines — stays where its author put it; the contents beside it link to each section on GitHub.

kordoc 작업 지침

kordoc — 한국 공문서(HWP 3.x·5.x, HWPX, HWPML, PDF, XLS·XLSX, DOCX, PPTX, 이미지)를 Markdown·IR 로 바꾸고 비교·생성하는 파서 라이브러리. npm 패키지로 배포하고 라이브러리 API·CLI(kordoc)·MCP 서버(kordoc-mcp, 도구 22개)를 낸다.

이 파일이 에이전트 공통 지침의 정본이다. Codex 는 이 파일을 바로 읽고, Claude Code 는 CLAUDE.md 의 @AGENTS.md 로 읽는다. 상세 참고: docs/architecture.md(모듈 구조·설계 결정·구현 주의 상세), docs/corpus.md(벤치 코퍼스 이력), docs/benchmarks.md(채점 기준·옵션별 수치), docs/gongmunseo-engine-spec.md(공문서 생성 엔진).

0. 최우선 규칙: 독자 구현

다른 모든 지침보다 앞섬 (2026-10-25 확정).

  1. 경쟁 제품(다른 한글 문서 편집기·오피스 앱) 저장소의 코드·패치·시험·문구·빌드 결과를 옮기지 않음. 라이선스가 허락해도 금지.
  2. 경쟁 제품은 기능 비교(무엇이 있나)까지만 봄. 구현을 보고 따라 짜지 않음. 구현 근거는 공개 명세(한컴 공식 도움말·HWP 공개 문서), rhwp 원본(MIT), kordoc 뿐.
  3. 엔진은 kordoc + npm rhwp 원본. rhwp 결함은 rhwp 에 PR 로 고침. 남의 패치 묶음을 들고 있지 않음.
  4. 예전에 옮겨 온 코드(NOTICE 에 적힌 것)는 우리 코드로 차례로 다시 짜고, 다 바꾸면 NOTICE 에서 뺌.
  5. 커밋 메시지·README·화면 문구에 경쟁 제품 이름 금지. 우리 프로젝트 이야기만 짧게.

작업 규칙

  • 사용자가 개선 완료와 반영을 요청한 작업은 검증 후 main에 반영하고 원격에 푸시한다. 이 저장소에서는 별도 PR이 필수 경로가 아니다. 병렬 작업과 충돌할 수 있으면 독립 worktree에서 준비한 뒤 최신 main을 확인하고 반영한다.
  • 배포까지 요청되었거나 사용자가 이 작업에 적용한 배포 관례를 명시했다면, 검증된 main에서 버전·메타·변경 내역을 맞춘 뒤 npm 게시와 GitHub 릴리스를 완료한다. 게이트 실패나 불명확한 게시 상태가 있으면 원인을 해결하고 이중 게시를 피한다.
  • 다른 작업 공간의 미커밋 파일과 코퍼스 심링크는 건드리거나 커밋하지 않는다. 관련 없는 PR의 변경은 섞지 않는다.
  • PDF 품질 작업에서는 정답·평가기·제외 모수를 바꾸어 점수를 높이지 않는다. 외부 ODL 200문서, 한국 공문서 PDF 표·글, HWPX/HWP5 및 자원 사용을 따로 검증한다.
  • 상위 공통 지침의 Git 정책과 다른 부분은 사용자가 이 저장소에 대해 2026-09-25에 명시한 main 반영·푸시·배포 관례를 따른다. 이후 사용자가 작업별로 범위를 제한하면 그 지시를 우선한다.
  • 채점기·정규화를 바꿔야 하면(새 출력 표지를 걷는 등) CHANGELOG 에 "채점 기준 변경"으로 적는다.

명령

npm run build          # tsup — ESM+CJS 라이브러리 + CLI·MCP 바이너리 → dist/
npm run dev            # watch 모드
npm test               # node --import tsx --test tests/*.test.ts
npm run typecheck      # tsc --noEmit
npm run bench:gate     # 코퍼스 회귀 게이트 체인(pages·score·roundtrip·pdf-table·pdf-text·formats·fuzz·reflow·redact·ocr·hwp2hwpx)
npm run bench:visual   # 선택 진단: 한컴 실렌더 비교 (macOS GUI 전용 — bench/visual/, 순수 로직은 hash-lib.mjs)
npm run sync-meta      # 공문서 엔진 스펙·구성 기준 정본(docs/) → 스킬 사본(.claude/skills/gongmunseo/references/) 동기화
node bench/predict-layout.mjs 문서.hwpx --loose   # 생성 공문서의 한글 조판 예측(실글꼴 폭표, 한글 2024 PDF 154줄 재현) — 한컴 없이 벌어진 줄·고아 줄·압축 확인

Read the full file on GitHub · 97 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 6d ago Changed · +6 tokens per session d866dff212ce
  2. 9d ago Changed 68fdcf13ebb0
  3. 11d ago Changed · +10 lines · +257 tokens per session 0a76e50a719f
  4. 16d ago Changed 808cd862097c
  5. 17d ago Changed · +14 tokens per session 9c6eeb5745d9
  6. 22d ago Changed · +1 lines · +58 tokens per session 23ec886c7c40
  7. 27d ago Changed · +1 lines · +82 tokens per session 3375dcee83a2
  8. 1mo ago Changed · +1 lines · +65 tokens per session 7fce59209a93
  9. 1mo ago Changed · +76 lines · +2,347 tokens per session 5f1d5642b888
  10. 1mo ago First seen · 8 lines · 330 tokens per session scan A 69d6334abe77

Subscribe to this mod's changes

kordoc AGENTS.md is an instructions file published in the GitHub repository chrisryugj/kordoc (2,928 stars, last pushed yesterday), licensed MIT. It adds 3,159 tokens to every session, about $0.0126 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-26.

Related

kordoc AGENTS.md compared with instructions from other repositories

kordoc AGENTS.md next to 6 instructions from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
kordoc AGENTS.mdchrisryugj/kordocInstructions file2.9kyesterdayMITA3,159Codex, OpenCode
vscode buildNext.instructions.mdmicrosoft/vscodeInstructions file195kyesterdayMITA9,719GitHub Copilot
next.js AGENTS.mdvercel/next.jsInstructions file143ktodayMITA7,461Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchainInstructions file149kyesterdayMITA4,680Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscodeInstructions file195kyesterdayMITA5,516GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscodeInstructions file195kyesterdayMITA3,093GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harnessInstructions file274k2d agoMITA4,313Codex, OpenCode