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 kimlawtech/korean-jangbu-for --skill jangbu-taggit clone --depth 1 https://github.com/kimlawtech/korean-jangbu-forWrote 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/kimlawtech/korean-jangbu-for/jangbu-tag)<a href="https://agentmods.dev/skills/kimlawtech/korean-jangbu-for/jangbu-tag"><img src="https://agentmods.dev/badge/skills/kimlawtech/korean-jangbu-for/jangbu-tag/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/kimlawtech/korean-jangbu-for/jangbu-tag"><img src="https://agentmods.dev/badge/skills/kimlawtech/korean-jangbu-for/jangbu-tag.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 | $0.00080 | $0.01645 |
| Opus 5 | $0.00040 | $0.00822 |
| Sonnet 5 | $0.00016 | $0.00329 |
| Haiku 4.5 | $0.00008 | $0.00164 |
Grade A, and why
jangbu-tag 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 12d 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 — 166 lines — stays where its author put it; the contents beside it link to each section on GitHub.
jangbu-tag
거래내역 → 계정과목 매핑 스킬.
분류 전략
- 룰 기반 (LLM 미경유) —
classify_with_rules호출로 일괄 처리 - LLM fallback — 룰 실패 건에 한해 마스킹된 뷰만 LLM에 전달
- 사용자 확인 — 낮은 confidence 건은 사용자 검증 후
apply_classification(source="user")
계정 구조 (이중)
- 내부 계정: 유연. 예)
광고선전비_구글,광고선전비_메타,SaaS_노션 - 국세청 계정: 고정.
accounts/nts_standard.json매핑 테이블 경유 - 세무 리포트 생성 시 내부 → 국세청 자동 변환
인터뷰 플로우
Step 1. 대상 기간 확인
분류할 거래 기간을 선택하세요.
[1] 이번 달
[2] 전월
[3] 올해 전체
[4] 특정 기간 입력
미분류 거래만 처리합니다.
Step 2. 룰 기반 일괄 분류
classify_with_rules(start_date="2026-04-01", end_date="2026-04-30")
결과 예시:
룰 분류 완료: 87건
분류 실패(LLM fallback 필요): 13건
Step 3. LLM fallback (분류 실패 건)
핵심: list_transactions(unclassified_only=True) 응답은 이미 마스킹된 뷰.
LLM 분류 프롬프트 템플릿
다음 거래를 한국 일반기업회계기준 계정과목으로 분류하세요.
거래처명·적요에 토큰(TK_*)이 있으면 그대로 두고 문맥으로만 판단.
거래 목록:
{
"tx_abc": {"date": "2025-09-07", "amount": 142025, "direction": "outflow",
"counterparty": "CLAUDE.AI SUBSCRIPTION", "description": "..."},
"tx_def": {...}
}
사용 가능 계정과목 (이 목록 외 사용 금지):
- 복리후생비: 직원 복리·회식·식대·간식·건강검진
- 여비교통비: 택시·KTX·주유·출장
- 접대비: 고객 접대 (5만원 이상 식사가 유력 신호)
- 통신비: 휴대폰·인터넷·유선전화
- 수도광열비/전력비: 수도·전기·가스
- 임차료: 월세·사무실 임대
- 지급수수료: SaaS 구독·은행 수수료·법무·회계·세무 용역비
- 광고선전비: 페이스북·구글·네이버 광고
- 도서인쇄비: 책·출판물·인쇄
- 소모품비: 사무용품·다이소·IKEA
- 세금과공과: 4대보험·세금·지방세
- 수선비·보험료·차량유지비·운반비·교육훈련비·회의비
- 급여·상여금·퇴직급여·잡급
- 매출액/이자수익 (inflow일 때)
각 거래에 대해 JSON으로 반환:
{
"tx_abc": {"internal_account": "지급수수료", "confidence": 0.92,
"reason": "CLAUDE.AI는 AI SaaS 구독"},
...
}
confidence < 0.7인 건은 "사용자 확인 필요"로 표시.
적용 절차
list_transactions(unclassified_only=True, limit=50)— 50건씩 배치- 위 프롬프트로 LLM 호출 → JSON 결과
- confidence ≥ 0.7:
apply_classification(source="llm")자동 저장 - confidence < 0.7: Step 4로 사용자 확인
LLM 분류 시 반드시 한국 일반기업회계기준의 표준 계정 사용.
Step 4. 낮은 confidence 건 사용자 확인
다음 거래의 계정을 확인해주세요.
거래: TK_ACCOUNT_ID_xxxx | 2026-04-15 | 500,000원 | outflow
적요: (마스킹된 내용)
추정 계정: 지급수수료 (confidence=0.65)
[1] 지급수수료 유지
[2] 다른 계정 지정
[3] 스킵 (미분류 유지)
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.
- 12d ago First seen · 166 lines · 80 tokens per session scan A 545a18fda02e
jangbu-tag is a skill published in the GitHub repository kimlawtech/korean-jangbu-for (83 stars, last pushed 13d ago), licensed Apache-2.0. It adds 80 tokens to every session and 1,645 once invoked, about $0.0004 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
bookkeeping
Use when a small business needs audit-ready books — a chart of accounts, posting a transaction to the right account and side, clearing an uncategorized bank feed, cash vs accrual, or a ledger that won't tie to the bank. NOT interpreting the numbers — runway, burn, P&L cadence (that is finance-ops), NOT issuing…
countbean-accounting
Rules for writing correct double-entry Beancount transactions when managing a Countbean cloud book — account naming, how to balance postings, categorisation, and safe ingestion. Use whenever recording, importing, or correcting financial data in a Countbean book.
ai-anomaly-detection
When the user wants to use machine learning to detect fraud, errors, or unusual patterns in high-volume financial data. Also use when the user mentions "ML fraud detection," "unsupervised learning for audit," "isolation forest," "autoencoders for finance," "unusual transaction clusters," or "automated expense…
bukio-cli
Drive bukio-cli — agent-first double-entry bookkeeping for SMEs across eleven jurisdictions. Book invoices, archive the originals, keep the books balanced.
QuickBooks Automation
Automate QuickBooks accounting workflows including invoicing, expenses, reporting, and bank reconciliation.
ledger
Use for ANY monetary, accounting, ledger, journal, balance, P&L, fee, FX, invoice, tax, cost-basis, or valuation work. Exact Money.from (never floats or parseFloat), double-entry via JournalEntry + validateEntry + Ledger.apply, MCP moneycompute / ledgerpost, canon citations, Zero-Skip + Canonical Financial Artifact.…