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 cosmix/loom --skill loom-i18ngit clone --depth 1 https://github.com/cosmix/loomWrote 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/cosmix/loom/loom-i18n)<a href="https://agentmods.dev/skills/cosmix/loom/loom-i18n"><img src="https://agentmods.dev/badge/skills/cosmix/loom/loom-i18n/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/cosmix/loom/loom-i18n"><img src="https://agentmods.dev/badge/skills/cosmix/loom/loom-i18n.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 | $0.00015 | $0.03651 |
| Opus 5 | $0.00008 | $0.01826 |
| Sonnet 5 | $0.00003 | $0.00730 |
| Haiku 4.5 | $0.00002 | $0.00365 |
Grade C, and why
loom-i18n scanned grade C with 1 finding 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 5d 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.
Hidden instructionshighPrompt injection
Directives inside HTML comments, invisible characters or bidirectional overrides are read by the model and not by the person reviewing the file.
- **Bidi isolation**: user-generated or opposite-direction text embedded in a sentence (an Arabic name in English UI, a phone number) can reorder surrounding punctuation. Wrap it in `<bdi>…</bdi>`, or `unicode-bidi: isol How it starts
The opening of the file, as written. The whole thing — 294 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Internationalization (i18n)
Overview
Designing software to adapt to languages/regions without code changes (i18n), then adapting it per-locale (l10n). Covers translation architecture, ICU pluralization, Intl-based formatting, RTL/bidi, and libraries (i18next, react-intl/FormatJS, gettext).
First: Check for an L10N.md Charter
Before writing or translating any string, check the project root for L10N.md. When it exists, it is the project's authoritative localization charter — read it in full and let it override the generic guidance in this skill wherever the two conflict. Derive from it:
- Project context — the domain the product operates in (financial, technological, medical, …). Domain dictates terminology register and tone: "credit" means one thing in a banking app and another in a game, and a payments product cannot afford a casual mistranslation of a regulated term.
- What to translate directly and what not — the charter's do-not-translate list: brand and product names, trademarks, legal or regulated terms, technical identifiers. Never translate an entry on that list, even when a natural target-language equivalent exists.
- Branding and glossary — approved per-locale renderings of recurring product terms. Reuse them verbatim; inventing a second translation for an established term fragments the product's voice across locales.
Only the root L10N.md is read. Some projects nest per-area charters (server/L10N.md) or per-locale overrides (L10N/es.md); those are not consulted here, so anything that must always apply belongs in the root file.
When L10N.md does not exist, skip this step entirely — proceed with the generic guidance below, and do not create the file unasked.
The Rules That Prevent Rework
-
Never concatenate translated fragments. Word order, gender agreement, and grammar differ per language. Use one full-sentence key with named placeholders; the translator controls order.
// Wrong — impossible to translate; order is baked into code t("You have") + " " + count + " " + t("new messages"); // Right — one message, interpolation + plural inside it t("inbox.newMessages", { count }); // "{count, plural, one {# new message} other {# new messages}}"
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.
- 5d ago Changed · +18 lines 2627e32248ca
- 6d ago Changed · -42 tokens per session a13dddfc7783
- 9d ago First seen · 276 lines · 57 tokens per session scan C fad684afafd9
loom-i18n is a skill published in the GitHub repository cosmix/loom (54 stars, last pushed yesterday), licensed MIT. It adds 15 tokens to every session and 3,651 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it C with 1 finding (hidden instructions). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
localized-pricing
Guide for implementing localized pricing, adaptive currency, and purchasing power parity with Dodo Payments.
API Discoverability for Agents
Making self-hosted services agent-discoverable — bake in a machine-readable API description (OpenAPI spec or a minimal API.md) when building, and discover-first (spec paths, repo search) before probing when integrating.
Agent Design Principles
A checklist for designing agent personas, skills, and multi-agent pipelines that stay reliable as they grow — grounded in the 12-factor-agents principles.
Internationalizing Websites
Adds multi-language support to Next.js websites with proper SEO configuration including hreflang tags, localized sitemaps, and language-specific content. Use when adding new languages, setting up i18n, optimizing for international SEO, or when user mentions localization, translation, multi-language, or specific…
dev-spec-driven-implementation
Use to implement a scoped multi-step change from an existing spec, handoff, issue, or explicit delivery request. Do not trigger for read-only analysis, planning, or a trivial localized edit.
dev-bugfix
Use for a concrete defect or regression that needs reproduction, root-cause diagnosis, a fix, and regression validation. Do not trigger for routine localized edits or primarily performance-focused work.