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 MSApps-Mobile/claude-plugins --skill rtl-chatgit clone --depth 1 https://github.com/MSApps-Mobile/claude-pluginsWrote 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/msapps-mobile/claude-plugins/rtl-chat)<a href="https://agentmods.dev/skills/msapps-mobile/claude-plugins/rtl-chat"><img src="https://agentmods.dev/badge/skills/msapps-mobile/claude-plugins/rtl-chat/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/msapps-mobile/claude-plugins/rtl-chat"><img src="https://agentmods.dev/badge/skills/msapps-mobile/claude-plugins/rtl-chat.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00126 | $0.02182 |
| Opus 5.5 | $0.00050 | $0.00873 |
| Sonnet 5.5 | $0.00025 | $0.00436 |
| Haiku 4.5 | $0.00013 | $0.00218 |
Grade A, and why
rtl-chat 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 1mo 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 — 221 lines — stays where its author put it; the contents beside it link to each section on GitHub.
RTL Chat Formatting
You are responding to a user who writes in an RTL (right-to-left) language in a chat interface (Cowork / Claude Desktop) that has LIMITED RTL support. The renderer right-aligns RTL text but struggles with word ordering when RTL and LTR scripts mix on the same line. Structure responses so they are as readable as possible given this constraint.
Supported RTL Languages
This skill applies when the user writes in any of these languages/scripts:
Hebrew (עברית), Arabic (العربية), Persian/Farsi (فارسی), Urdu (اردو), Pashto (پښتو), Sindhi (سنڌي), Kurdish (Sorani: کوردی), Yiddish (ייִדיש), Dhivehi/Maldivian (ދިވެހި)
Detect the language from the user's input and respond in the same language. All formatting rules below apply equally to every RTL language.
The Core Problem and Strategy
When RTL and LTR text appear on the same line, the BiDi (bidirectional) algorithm often scrambles the word order. The more LTR terms on a single line, the worse it gets. The strategy: minimize mixing on any single line.
DO NOT use Unicode directional marks (RLM, LRM, RLE, etc.). They do not improve rendering in this environment and can make things worse.
Rules for Readable Mixed Text
Rule 1: One LTR term per line, maximum
The most important rule. Each line should contain at most ONE English/LTR term or phrase. If multiple LTR terms are needed, split them across separate lines with a blank line between each.
Bad — multiple LTR terms on one line:
לניהול סטייט אני ממליץ על Zustand או Redux Toolkit עם TypeScript
لإدارة الحالة أنصح باستخدام Zustand أو Redux Toolkit مع TypeScript
Good — one term per line:
לניהול סטייט אני ממליץ על Zustand
אפשרות נוספת היא Redux Toolkit
שניהם עובדים מצוין עם TypeScript
لإدارة الحالة أنصح باستخدام Zustand
خيار آخر هو Redux Toolkit
كلاهما يعمل بشكل ممتاز مع TypeScript
Rule 2: Put LTR terms at the END of the line
Structure the RTL sentence so the English/LTR term appears at the end. The BiDi algorithm handles trailing LTR text better.
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.
- 1mo ago First seen · 221 lines · 126 tokens per session scan A 2fcd858e7f57
rtl-chat is a skill published in the GitHub repository MSApps-Mobile/claude-plugins (10 stars, last pushed today), licensed MIT. It adds 126 tokens to every session and 2,182 once invoked, about $0.0005 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 2026-09-13.
Other skills, from other repositories
i18n-check
An internationalization (i18n) check for software that supports multiple languages. It compares translation files with the text and translation keys used by the code.
seo-hreflang
Hreflang and international SEO audit, validation, and generation. Detects common mistakes, validates language/region codes, and generates correct hreflang implementations for HTML, HTTP headers, and XML sitemaps. Use when user says "hreflang", "i18n SEO", "international SEO", "multi-language", "multi-region"…
i18n-l10n
Use when a ticket adds user-facing text or must support more than one language, locale, or region — translations, plurals, dates and numbers, currencies, right-to-left layout, time zones — and the implementation must externalise every string, use ICU/CLDR plural rules, format by locale with the platform's APIs, and…
review
5-pass structured code review — correctness, security, performance, readability, consistency.
wiki
Markdown-first knowledge base where the LLM acts as librarian. Ingests raw sources, compiles and interlinks topic files, self-maintains an index. No vector DB or embeddings required -- uses LLM-native navigation over structured markdown up to 400K words.
marshal
Meta-orchestrator that takes any direction — broad, specific, or vague — and autonomously chains skills and context into actionable work. Gathers context from codebase, docs, and memory. Only asks the user when it genuinely cannot proceed. Single-session orchestrator.