Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add JangHyun-bin/korean-report-skills --skill korean-report-docgit clone --depth 1 https://github.com/JangHyun-bin/korean-report-skillsWrote 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.
[](https://agentmods.dev/skills/janghyun-bin/korean-report-skills/korean-report-doc)<a href="https://agentmods.dev/skills/janghyun-bin/korean-report-skills/korean-report-doc"><img src="https://agentmods.dev/badge/skills/janghyun-bin/korean-report-skills/korean-report-doc/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.
<a href="https://agentmods.dev/skills/janghyun-bin/korean-report-skills/korean-report-doc"><img src="https://agentmods.dev/badge/skills/janghyun-bin/korean-report-skills/korean-report-doc.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00135 | $0.03284 |
| Opus 5 | $0.00068 | $0.01642 |
| Sonnet 5 | $0.00027 | $0.00657 |
| Haiku 4.5 | $0.00014 | $0.00328 |
Grade A, and why
korean-report-doc 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 13d 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.
How it starts
The opening of the file, as written. The whole thing — 249 lines — stays where its author put it; the contents beside it link to each section on GitHub.
한국어 기술 문서 제작
한국어 기술·사업 문서를 자립형 HTML로 제작하고 필요하면 PDF로 출력한다.
- 두 출력 모드는 공통 디자인 token과 모드별 CSS를 사용한다.
- 수치·표·도해는 제공된 데이터에서 프로그램으로 생성한다.
- HTML은 실행 시점에 외부 자원을 요청하지 않는다.
- 본문 글꼴은
--font로 지정한 경우에만 내장된다.
문장 규약은 korean-report-style 스킬이 담당한다. 이 스킬의 범위는 제작에 한정한다.
1. 시작 전 결정
1.1 모드
| paper | deck | |
|---|---|---|
| 용도 | 분석 · 벤치마크 · 기술보고 · 진행현황 | 회의 · 아젠다 · 의사결정 요청 |
| 방향 | 세로 | 가로 |
| 페이지 | 연속 흐름 | 1 섹션 = 1 페이지 |
| 본문 폭 | 720px | 1120px |
| 밀도 | 산문 전체 | 한 페이지에 한 가지 |
| 템플릿 | assets/paper_template.html |
assets/deck_template.html |
| 본문 구조 | <div class="doc"> 안의 <section> |
<section class="tile ..."> 나열 |
이 문서에서 layout은 페이지와 요소의 배치를, typesetting은 CSS·글꼴·수식을
반영해 최종 HTML을 생성하는 처리 단계를 뜻한다.
섞지 않는다. 문단이 긴 deck 은 deck의 layout이 어긋난 보고서로 읽히고,
타일 배경을 쓴 paper 는 슬라이드를 잘못 인쇄한 것으로 읽힌다.
1.2 산출물 필요 여부
문서 산출물을 요청받았을 때만 작성한다. 설명·요약·분석은 대화로 답한다.
보고서 만들어줘 · PDF로 뽑아줘 · 문서로 정리해줘 가 신호다.
1.3 실행 환경
- HTML 빌드에는 Node 20 이상과 KaTeX가 필요하다.
mathbuild.js는 작업 디렉터리와 스킬 설치 경로에서 KaTeX를 순서대로 찾는다. - 렌더 QA, 스크린샷, PDF 출력에는 Python 3.11 이상, Playwright, Chromium이 필요하다.
- 본문 글꼴을 내장하려면 사용 권한이 있는 WOFF2 파일을 준비한다.
2. 빌드 파이프라인
데이터 ──▶ 생성기(Python) ──▶ raw HTML (수식 마커 + 치환 토큰)
│
▼
후처리(Node + KaTeX) — mathbuild.js
· ⟦I⟧ / ⟦D⟧ 렌더
· base.css + 모드 CSS 삽입
· 지정한 본문 글꼴 · KaTeX woff2 base64 내장
│
▼
단일 .html ──▶ headless print ──▶ .pdf
템플릿은 치환 토큰만 담은 얇은 껍데기다. 실제 규칙은 assets/css/ 에 있다.
| 토큰 | 채우는 주체 |
|---|---|
__TITLE__ · __BODY__ |
생성기 |
__FONTCSS__ · __KATEXCSS__ · __BASECSS__ · __MODECSS__ |
mathbuild.js |
CSS 를 템플릿에 직접 쓰지 않는다. 두 템플릿이 규칙을 각자 복사해 두었다가
deck 의 인쇄 규칙이 paper 의 것에 덮인 사고가 있었다. 공통은 css/base.css,
모드별은 css/paper.css · css/deck.css 한 곳에만 둔다.
What ships with it
10 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- assets/css/base.css 12 KB
- assets/css/deck.css 3.7 KB
- assets/css/paper.css 2.5 KB
- assets/deck_template.html 1.2 KB
- assets/figures.py 18 KB runs code
- assets/mathbuild.js 9.5 KB runs code
- assets/paper_template.html 490 B
- assets/qa.py 6.6 KB runs code
- references/design.md 30 KB
- references/figures.md 7.9 KB
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.
- 13d ago First seen · 249 lines · 135 tokens per session scan A 5c6059d29e22
korean-report-doc is a skill published in the GitHub repository JangHyun-bin/korean-report-skills (74 stars, last pushed 8d ago), licensed Apache-2.0. It adds 135 tokens to every session and 3,284 once invoked, about $0.0007 per session on Opus 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 2026-08-30.
Other skills, from other repositories
kordoc
A command-line and server tool for reading and creating Korean official documents, including HWP, HWPX, PDF, DOCX, and spreadsheet files. It can convert documents to Markdown and create or update HWPX files.
book-pdf
A document-to-book publishing workflow that turns Markdown—a plain-text writing format—into book-style PDF and EPUB files. It includes book pages such as a cover, copyright page, numbered contents, chapter breaks, running headers, and page numbers.
docs-crawler
Crawl an entire documentation or design-system website into one LLM-ready Markdown corpus. Discovers every page from the site's sitemap.xml (with a same-origin link-following fallback when there is no sitemap), extracts each page's main content as clean Markdown, downloads each page's images (and inline base64 images)…
fullbleed
Use Fullbleed PDF Engine to create or modify document-generation projects for PDFs, reports, invoices, statements, forms, certificates, letters, accessible PDF/PDF-UA, archival PDF/PDF-A, print-ready PDF/PDF-X, transactional output, and fixed or reflowing variable-data/VDP jobs from structured content using Python and…
archforge-pptx-lint
Use when building, editing, or reviewing .pptx files - run the archforge linter on the built file before delivery, read its gate codes, fix defects, and re-lint until clean. Catches silent font fallback (Hangul-deep, CJK-aware), tracking damage, unreadable sizes, text collisions, off-canvas bleed, and AI-generated…
travel-guidebook
A workflow for creating designed travel guidebook PDFs from trip itineraries. It researches destinations, structures the route, writes the content, builds an HTML guide, checks it, and exports the result as a PDF.