Set Claude's Response Language with the language Setting
If you work in a language other than English, the language setting in settings.json tells Claude to respond in that language by default across all your sessions.
// ~/.claude/settings.json
{
"language": "japanese"
}
Supported values are plain language names: "japanese", "spanish", "french", "german", "portuguese", "korean", "chinese", and so on. Claude treats this as a strong preference, so you can still ask questions in English and Claude will reply in the configured language.
The setting also affects push-to-talk voice dictation. When language is set and you enable /voice, the dictation engine is configured to expect that language, which improves transcription accuracy for non-English speech.
You can set this globally in user settings for a personal preference, or in project settings if the codebase documentation is written in a specific language and you want Claude to match it consistently:
// .claude/settings.json (for a French-language project)
{
"language": "french"
}
There is no need to remind Claude of your language preference at the start of every session. Set it once and it applies everywhere.
One line removes the overhead of translating Claude's responses yourself.
Log in to leave a comment.
When Claude writes error messages, button labels, validation text, or onboarding flows, it defaults to generic developer-speak. Add a "Users" section to your CLAUDE.md describing who your actual users are — their technical level, industry jargon, and what they care about — so Claude writes copy that makes sense to THEM, not to developers.
Use the --agent flag with custom markdown files in .claude/agents/ to launch purpose-built Claude sessions with restricted tools and scoped system prompts.
Every project has traps — the billing module that silently fails if you forget to queue the job, the legacy table with column names that don't match the model, the config value that must be set before tests run. Document these gotchas in your CLAUDE.md so Claude avoids the same mistakes your team spent days debugging.