How do I install the I18N Context skill?

Run npx skillmds@latest add cataclysmbn/i18n-context in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is I18N Context safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with I18N Context?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is I18N Context free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published I18N Context?

cataclysmbn (@cataclysmbn) published this skill. Their other Agent Skills are listed on their SkillMD profile.

I18N Context by cataclysmbn — Agent Skill

Name: I18N Context
Author: cataclysmbn

File contents

I18n Context

Use this when one English msgid has multiple meanings and a locale needs different translations.

Hard requirements

Do not reduce requested word/context coverage for review risk, churn, or convenience.
If a requested meaning needs loader support for contextual translation, add loader support.
Do not open or update a PR while msgfmt or printf checks fail for the touched locale.
Do not dismiss msgfmt/printf failures as pre-existing when this task touches that locale.

Workflow

Find every shared msgid before editing:
- rg -n 'msgid "<word>"|msgctxt .*\nmsgid "<word>"' lang/po/<lang>.po
- rg -n '"<word>"|translate_marker\( "<word>" \)|_\( "<word>" \)' data src
Split meanings at the source, following #8729 and #8741:
- C++ runtime strings: pgettext( "short specific context", "text" ).
- C++ marker-only strings: translate_marker_context( "short specific context", "text" ).
- JSON translated strings: { "ctxt": "short specific context", "str": "text" } or matching plural fields.
- Snippet arrays: { "text": { "ctxt": "short specific context", "str": "text" } }.
- If a JSON field is currently std::string, convert it to translation so contextual JSON can load.
Keep context names terse and UI-specific, e.g. time format, map extra, music genre, damage type.
Update lang/po/<lang>.po for each new msgctxt; do not rely on the old shared msgstr.
Preserve old uncontexted entries only when some source still uses the shared msgid.
For Korean, check existing glossary first:
- rg -C2 -i '<term>' lang/po/ko.po | rg -v '^(#:|--)' | head -n 20
Validate after editing:
- python3 lang/extract_json_strings.py -i data/json -o /tmp/catabn-i18n-context.pot --tracked-only
- msgfmt -f -c -o /tmp/ko.mo lang/po/ko.po
- ./tools/check_po_printf_format.py
- If JSON changed, run ./build-scripts/lint-json.sh.

PR references

#8729 split ambiguous yes/no labels in C++ with pgettext and translate_marker_context.
#8741 split ambiguous animal/action/species/fish names in JSON with ctxt and updated PO entries.

---name: i18n-contextdescription: Add gettext context for ambiguous user-facing strings and repair affected PO translations.---# I18n ContextUse this when one English msgid has multiple meanings and a locale needs different translations.## Hard requirements- Do not reduce requested word/context coverage for review risk, churn, or convenience.- If a requested meaning needs loader support for contextual translation, add loader support.- Do not open or update a PR while `msgfmt` or printf checks fail for the touched locale.- Do not dismiss `msgfmt`/printf failures as pre-existing when this task touches that locale.## Workflow- Find every shared msgid before editing: - `rg -n 'msgid "<word>"|msgctxt .*\nmsgid "<word>"' lang/po/<lang>.po` - `rg -n '"<word>"|translate_marker\( "<word>" \)|_\( "<word>" \)' data src`- Split meanings at the source, following #8729 and #8741: - C++ runtime strings: `pgettext( "short specific context", "text" )`. - C++ marker-only strings: `translate_marker_context( "short specific context", "text" )`. - JSON translated strings: `{ "ctxt": "short specific context", "str": "text" }` or matching plural fields. - Snippet arrays: `{ "text": { "ctxt": "short specific context", "str": "text" } }`. - If a JSON field is currently `std::string`, convert it to `translation` so contextual JSON can load.- Keep context names terse and UI-specific, e.g. `time format`, `map extra`, `music genre`, `damage type`.- Update `lang/po/<lang>.po` for each new `msgctxt`; do not rely on the old shared `msgstr`.- Preserve old uncontexted entries only when some source still uses the shared msgid.- For Korean, check existing glossary first: - `rg -C2 -i '<term>' lang/po/ko.po | rg -v '^(#:|--)' | head -n 20`- Validate after editing: - `python3 lang/extract_json_strings.py -i data/json -o /tmp/catabn-i18n-context.pot --tracked-only` - `msgfmt -f -c -o /tmp/ko.mo lang/po/ko.po` - `./tools/check_po_printf_format.py` - If JSON changed, run `./build-scripts/lint-json.sh`.## PR references- #8729 split ambiguous yes/no labels in C++ with `pgettext` and `translate_marker_context`.- #8741 split ambiguous animal/action/species/fish names in JSON with `ctxt` and updated PO entries.

Frequently asked questions

I18N Context