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 sangrokjung/claude-forge --skill humanize-koreangit clone --depth 1 https://github.com/sangrokjung/claude-forgeWrote 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/sangrokjung/claude-forge/humanize-korean)<a href="https://agentmods.dev/skills/sangrokjung/claude-forge/humanize-korean"><img src="https://agentmods.dev/badge/skills/sangrokjung/claude-forge/humanize-korean/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/sangrokjung/claude-forge/humanize-korean"><img src="https://agentmods.dev/badge/skills/sangrokjung/claude-forge/humanize-korean.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 | $0.00145 | $0.01295 |
| Opus 5 | $0.00072 | $0.00647 |
| Sonnet 5 | $0.00029 | $0.00259 |
| Haiku 4.5 | $0.00015 | $0.00129 |
Grade A, and why
humanize-korean 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.
How it starts
The opening of the file, as written. The whole thing — 46 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Humanize Korean — single-pass cleanup skill (한글 윤문 스킬)
Analogy. A proofreading desk for text that reads like it just came out of a translator — the goal is to make it read like it was written in Korean from the start. It changes not a single fact (proper nouns, numbers, quotes stay 100% intact) — only the style and rhythm. This is removing translation-ese and AI idioms to improve Korean prose quality, not "evading an AI detector."
비유: 번역기를 갓 거친 듯 어색한 한글을 "원래 한국어로 쓴 글"처럼 다듬는 교정 데스크. 사실은 한 글자도 안 바꾸고(고유명사·수치·인용 100% 보존) 문체·리듬·표현만 손본다. 이것은 번역투·AI 구문 제거 = 한글 품질 개선이지 "AI 탐지기 우회"가 아니다.
0. Before you start — scope and data-surface check
- Routing check. If the input is marketing copy that hasn't been written yet, don't run this skill on it — write it clean the first time per
rules/korean-writing-quality.mdinstead; this skill is for cleaning up existing text. If the input is a full book manuscript, treat this as a starting point only — long-form editorial judgment belongs to a human editor or a project-specific process. - Sensitive-text check (see
rules/korean-writing-quality.md§5). If this session or your tooling routes the actual rewrite through a model surface outside this session (a different vendor's API, a hosted service), don't send text containing client names, deal amounts, PII, salary data, or secrets through that surface unless you've explicitly cleared it for that data. When in doubt, do the cleanup pass locally in this session, or redact first.
1. Single-pass workflow (in this session, one pass)
- Read the input. If it's long (roughly 5,000+ Korean characters), consider splitting it into logical sections and running the pass per section — don't try to hold the whole self-check in your head for a very long piece at once.
- Detect. Scan the text against
references/quick-rules.md(compact ID + prescription table) and, for anything ambiguous or worth a worked example,reference/ai-tell-taxonomy.md(the full pattern reference this pack ships at the repo root). - Rewrite. Fix only the spans you detected — a surgical edit, not a rewrite of the whole piece. Keep every proper noun, number, date, unit, and direct quote byte-for-byte identical.
- Self-check. Run the six-item checklist in
references/quick-rules.mdbefore returning the result. If change rate exceeds 50%, stop and tell the user instead of shipping the rewrite — that's not a cleanup pass anymore. - Return. The cleaned text, plus: (a) a one-line status (change rate estimate, rough grade A-D per
references/quick-rules.md), and (b) 3-6 of the most significant before → after examples so the user can see what actually changed.
What ships with it
1 file 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.
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.
- 6d ago First seen · 46 lines · 145 tokens per session scan A 82c001510861
humanize-korean is a skill published in the GitHub repository sangrokjung/claude-forge (835 stars, last pushed 7d ago), licensed MIT. It adds 145 tokens to every session and 1,295 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 2026-09-03.
Other skills, from other repositories
i18n-localization
Internationalization and localization for global applications. Use when adding multi-language support, handling regional formats, or preparing apps for global markets.
stock-analysis-lead
Orchestrate a US-stock investment analysis — classify sector archetype, fetch SEC filings, dispatch a tiered fan-out of six vertical equity-research agents (business model, earnings quality, balance sheet, management, industry, peer comparison) over a validated JSON findings contract, then synthesize a buy/hold/sell…
api-design
REST API contract designer and reviewer. ALWAYS use when designing new endpoints, reviewing existing API contracts, planning API versioning, or standardizing error models. Covers resource modeling (URL/naming), HTTP method semantics, status code selection, error model consistency, pagination/filtering/sorting…
kafka-event-driven-design
Kafka event-driven architecture designer and reviewer, at the application/client layer. ALWAYS use when designing, reviewing, or troubleshooting how a service produces or consumes Kafka events — topic and partition-key design, producer and consumer client configuration, consumer group topology, event schema definition…
log-analyzer
Senior-SRE log analysis specialist. Use when investigating incidents from logs, triaging error spikes, extracting timelines, correlating distributed traces, or separating signal from noise across plain-text, JSON (slog/zap), syslog, journald, container, and Kubernetes logs. ALWAYS use when the user asks to "analyze…
tech-doc-writer
A guide for writing, reviewing, and improving technical documents such as API documentation, design documents, runbooks, and incident reports.