What it does
Reads a translated piece of content alongside a note on the target locale and flags spots where an idiom, cultural reference, image, or example was translated literally but won't make sense or could land badly for that audience, then suggests a locally resonant alternative. Focuses specifically on cultural fit, not grammar or literal translation accuracy, which is assumed already handled.
Install
This skill isn't packaged in an installable registry. It's a plain SKILL.md file you can copy directly. Copy it from the SKILL.md tab and save it at the path for your assistant:
Use it in your AI assistant
Claude Code
.claude/skills/cultural-localization-reviewer/ (or ~/.claude/skills/cultural-localization-reviewer/ for all projects)
GitHub Copilot (VS Code)
.github/skills/cultural-localization-reviewer/
OpenAI Codex
.codex/skills/cultural-localization-reviewer/
Cursor
.cursor/skills/cultural-localization-reviewer/ (or ~/.cursor/skills/cultural-localization-reviewer/ globally)
Compatibility
Any WordPress site with translated content in place, via WPML, Polylang, TranslatePress, or manual translation; works from the translated text plus target locale as input.
The complete SKILL.md for this skill:
---
name: cultural-localization-reviewer
description: Use when a user wants translated content reviewed for cultural fit, or asks whether an idiom or reference will "land" in a target locale, or for a "localization review".
---
# Cultural Localization Reviewer
## When to use
Content has already been translated and the user wants a check on whether idioms, jokes, or references will actually resonate in the target locale, separate from a grammar or accuracy check.
## Procedure
1. Read the translated content in full alongside the target locale or region it's intended for.
2. Flag any idiom, joke, or figure of speech that was translated literally but likely won't carry its meaning in that locale.
3. Flag references such as holidays, sports, celebrities, measurement units, or currency that are specific to the source culture and won't resonate or make sense locally.
4. For each flagged item, suggest a locally resonant alternative or note that it should simply be cut if no good equivalent exists.
5. Note anything that could be perceived as insensitive or awkward in the target culture even if it was harmless in the source.
## Guardrails
- This is a cultural-fit review only, not a grammar or translation-accuracy check, defer to the translator or a native speaker for those separately.
- Don't assume you know a locale's sensitivities better than a native speaker, flag for human judgment rather than asserting a rewrite is definitely correct.