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 agentmods add skills/kouroshez/coding-os/i18nnpx skills add kouroshez/coding-os --skill i18ngit clone --depth 1 https://github.com/kouroshez/coding-osWrote 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/kouroshez/coding-os/i18n)<a href="https://agentmods.dev/skills/kouroshez/coding-os/i18n"><img src="https://agentmods.dev/badge/skills/kouroshez/coding-os/i18n.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00212 | $0.03080 |
| Opus 5 | $0.00106 | $0.01540 |
| Sonnet 5 | $0.00042 | $0.00616 |
| Haiku 4.5 | $0.00021 | $0.00308 |
Grade A, and why
i18n 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 4d 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 — 181 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Internationalization & Localization — One Codebase, Every Locale
A practical guide to building software that adapts to language, region, and script without forking the code per market. Stack-agnostic; recipes target react-i18next / FormatJS (ICU) / next-intl / vue-i18n on the client and ICU / gettext on the server. The distinction that frames everything: i18n (internationalization) is the engineering to make software localizable; l10n (localization) is the per-locale content and adaptation that engineering enables.
When to Use This Skill
- Extracting hardcoded UI strings into a message catalog for the first time.
- Choosing an i18n library and message format (ICU MessageFormat vs gettext).
- Handling plurals/gender/grammar that differ by language (not every language has "1 vs many").
- Formatting dates, numbers, currency, and lists per locale.
- Supporting right-to-left (Arabic, Hebrew) and bidirectional text.
- Designing the locale-resolution chain (URL / header / cookie / user preference).
- Wiring the translator workflow — extraction, a translation-management system, re-import.
Skip when: the only concern is screen-reader/keyboard access for a single language — that is a11y. This skill is the language-and-locale dimension; the two compose (a localized UI must still be accessible).
Rule Zero — Never Concatenate Translated Strings
The defining beginner mistake: t("Hello") + " " + name + t("!") or building a sentence from fragments. Word order, agreement, and punctuation differ per language; concatenation hardcodes English grammar into every locale.
// WRONG — assumes English word order, breaks in most languages
const msg = t("You have") + " " + count + " " + t("new messages");
// RIGHT — one externalized message with named placeholders; the translator owns the whole sentence
t("inbox.unread", { count });
// catalog (ICU): "You have {count, plural, one {# new message} other {# new messages}}"
Every user-visible string is one complete, externalized message with named interpolation placeholders. The translator receives the full sentence and reorders freely.
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.
- 4d ago First seen · 181 lines · 212 tokens per session scan A b389dc0538d9
i18n is a skill published in the GitHub repository kouroshez/coding-os (6 stars, last pushed 4d ago), licensed Apache-2.0. It adds 212 tokens to every session and 3,080 once invoked, about $0.0011 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 2026-08-31.
Other skills, from other repositories
multilingual
Convert Hyper-Extract YAML templates to multi-language support. Use when user says: "add translation", "multilingual", "convert language", "support Chinese and English". Trigger: User wants to add Chinese/English/Japanese translation to their template. Skip: User wants to design a new template (use record-designer or…
doc-translator
将 PDF 格式或 Web Page 格式的源文档转换成保持原排版格式的 Markdown 格式精译中文文档。.
batch-processor
Manage large document processing by splitting into batches and coordinating the full translation workflow (extraction, translation, formatting). Use when handling documents larger than 30 pages, 60 paragraphs, or 6000 words.
web-translator
Extract web page content and convert to Markdown format while preserving structure, images, tables, and links. Use when converting web pages to Markdown for further processing.
zh-translator
Translate Markdown content from any language to Chinese while preserving formatting, code blocks, formulas, and special elements. Use when you need accurate Chinese translation of technical documents.
document-translate
将 Knowledge / Documents 文档的英文 Markdown 正文按段落分块高保真翻译为中文：代码块、行内代码、URL、图片路径、LaTeX 公式、HTML 标签、front-matter 键名逐字节保留不翻，Markdown 结构与原文一一对应，逐块翻译禁止合并/拆分/遗漏。Use when translating chunked Markdown documents (source/chunkNNNN.md → translated/chunkNNNN.md) in a translation workdir.