docs-i18nlisted

# docs-i18n — Internationalization Analysis ## Workflow 1. **Check language setup** — find which languages are enabled and which pages exist per language. If the docs are published on a platform with language settings (e.g. a Docsbook workspace), read them from there; otherwise infer from the folder structure (e.g. `docs/en`, `docs/fr`) or frontmatter. If only one language is in scope, skip this skill entirely. 2. **Gather the docs** — get the list of pages across all languages and read their content. If a semantic/graph search tool over the markdown is available (self-hosted `markdown-lsp`, or a connected Docsbook workspace), prefer it — faster and cheaper than scanning files; otherwise read the files directly with `grep`/`find`. 3. **Apply checklist** — check language configuration, content parity by tier, navigation/UI translation, translation content rules, format localization, RTL handling, translation freshness, and SEO for multilingual. 4. **Produce report** — return one JSON issue object per finding, sorted by severity. ## Guardrails - Do not run this skill if only one language is enabled — exit early with a note. - Confirm the source-of-truth language (usually English) with the user before flagging parity gaps. - Content parity is priority-based: Tier 1 pages must be translated; Tier 3 can stay English-only. - A stale translation is worse than no translation — flag staleness even if the page exists. - Code inside code blocks is never translated — only surrounding