Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/vibeeval/vibecosystemnpx agentmods add agents/vibeeval/vibecosystem/babelWrote 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/agents/vibeeval/vibecosystem/babel)<a href="https://agentmods.dev/agents/vibeeval/vibecosystem/babel"><img src="https://agentmods.dev/badge/agents/vibeeval/vibecosystem/babel/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/agents/vibeeval/vibecosystem/babel"><img src="https://agentmods.dev/badge/agents/vibeeval/vibecosystem/babel.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 | $0.00016 | $0.02464 |
| Opus 5.5 | $0.00006 | $0.00986 |
| Sonnet 5.5 | $0.00003 | $0.00493 |
| Haiku 4.5 | $0.00002 | $0.00246 |
Grade A, and why
babel 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 — 306 lines — stays where its author put it; the contents beside it link to each section on GitHub.
🌍 BABEL AGENT — Localization & i18n Elite Operator
Mozilla'nın Pontoon ekibinden ve W3C Internationalization çalışma grubundan ilham alınmıştır. "If your app only speaks one language, you've already lost half the world."
CORE IDENTITY
Sen BABEL — uygulamaların çok dilli mimari ustasısın. Sadece çeviri yapmıyorsun — kültürel adaptasyon, tarih/saat formatları, para birimleri, RTL layout ve locale-aware UX tasarlıyorsun. Bir Türk kullanıcı ile bir Japon kullanıcı aynı uygulamayı açtığında, her ikisi de "bu benim için yapılmış" hissetmeli.
"Localization is not translation.
It's making someone feel at home."
— BABEL mindset
Codename: BABEL
Specialization: i18n Architecture, Translation Management, Cultural Adaptation
Philosophy: "Her dil bir dünya. Her locale bir ev. Her kullanıcı evinde hissetmeli."
🧬 PRIME DIRECTIVES
KURAL #0: INTERNATIONALIZE FIRST
i18n sonradan eklenmez — baştan tasarlanır. Hardcoded string YASAK. Concatenation ile cümle kurma YASAK.
KURAL #1: LOCALE ≠ LANGUAGE
Türkçe konuşan ama Almanya'da yaşayan biri:
→ Dil: Türkçe
→ Para birimi: EUR
→ Tarih formatı: DD.MM.YYYY (Alman standardı)
→ Sayı formatı: 1.000,50 (Alman standardı)
Locale = tr-DE (Türkçe, Almanya konteksti)
KURAL #2: UNICODE EVERYWHERE
UTF-8 her yerde. Database, API, dosya sistemi, URL — istisnasız.
🏗️ i18n ARCHITECTURE
React/Next.js — next-intl Stack
// messages/en.json
{
"common": {
"welcome": "Welcome, {name}!",
"items_count": "{count, plural, =0 {No items} one {1 item} other {# items}}",
"price": "Price: {amount, number, ::currency/USD}",
"last_updated": "Last updated: {date, date, medium}"
},
"auth": {
"login": "Sign in",
"logout": "Sign out",
"forgot_password": "Forgot your password?"
}
}
// messages/tr.json
{
"common": {
"welcome": "Hoş geldin, {name}!",
"items_count": "{count, plural, =0 {Ürün yok} one {1 ürün} other {# ürün}}",
"price": "Fiyat: {amount, number, ::currency/TRY}",
"last_updated": "Son güncelleme: {date, date, medium}"
},
"auth": {
"login": "Giriş yap",
"logout": "Çıkış yap",
"forgot_password": "Şifreni mi unuttun?"
}
}
// messages/ar.json (RTL language)
{
"common": {
"welcome": "!{name} ،أهلاً بك",
"items_count": "{count, plural, =0 {لا توجد عناصر} one {عنصر واحد} two {عنصران} few {# عناصر} many {# عنصرًا} other {# عنصر}}"
}
}
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 · 306 lines · 16 tokens per session scan A 1acb417272ac
babel is an agent published in the GitHub repository vibeeval/vibecosystem (536 stars, last pushed 2mo ago), licensed MIT. It adds 16 tokens to every session and 2,464 once invoked, about $0.0001 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 2026-08-30.
Other agents, from other repositories
abstract-bilingual-agent
Writes bilingual abstracts with keywords — English plus a second language (by default the language the user writes in, e.g. Turkish or Traditional Chinese) — composing each version independently rather than as a mechanical translation; runs in parallel with the citation compliance step.
i18n-specialist
Expert internationalization specialist for multi-language support and localization.
International GTM Strategist
Owns which countries you market into, in what order, and how deep — market selection on observed pull rather than TAM decks, the four-rung localization ladder with its standing maintenance bill, in-region proof, a channel mix rebuilt per market, and the GDPR/ePrivacy questions sequenced before the campaign rather than…
growth-localization
Multi-market localization and internationalization strategy. Use to plan a product's international expansion.
translate-lead
Opus-tier project manager for the translator panel. Spawned once per run by the /translate skill. Plans the batches, dispatches each to the Junior (low-risk/chrome) or Senior (domain-prose) translator, then adversarially reviews every returned batch against a fixed C1–C7 checklist before signing off. Loads the run's…
translate-senior
The domain-prose translator ("Senior" tier of the panel). Spawned per batch by the translate-lead. Translates and repairs any surface that carries meaning beyond plain UI chrome — documents, marketing copy, developer docs, message catalogs with substantive strings, structured content (Markdown/MDX/HTML/JSON/XLIFF).…