[FEATURE] Permission prompt explanation (Ctrl+E) should respect language setting
Preflight Checklist
- [x] I have searched existing requests and this feature hasn't been requested yet
- [x] This is a single feature request (not multiple features)
Problem Statement
The permission prompt's "explanation" text (toggled via Ctrl+E) appears to be generated by a small LLM (likely Haiku-class), but it always responds in English regardless of the user's language setting configured via /config.
This is fundamentally different from the static UI localization issue tracked in #4866 (which covers hardcoded CLI strings like "Allow/Deny"). Since the explanation text is LLM-generated, it should be straightforward to pass the user's language preference to the model and generate the response in that language — no i18n framework or translation files needed.
Issue #23592 was closed as a duplicate of #4866, but it conflated two distinct problems:
- Static UI localization (button labels, status messages) — requires i18n infrastructure (#4866)
- LLM-generated explanation text — only requires passing the language setting to the model prompt
This issue specifically addresses problem (2).
Proposed Solution
When generating the permission explanation text (Ctrl+E), include the user's configured language (from /config → Set Language) in the prompt to the explanation model, so the response is generated in the user's preferred language.
Alternative Solutions
- Users can read the English explanation and mentally translate, but this defeats the purpose of the language setting.
- The current workaround is to just accept/deny without understanding the explanation.
Priority
Medium - Would be very helpful
Feature Category
Interactive mode (TUI)
Use Case Example
- User sets language to 日本語 via
/config - Claude generates a Bash command like
find /path | xargs rm - Permission prompt appears; user presses Ctrl+E to understand what the command does
- Currently: explanation is in English
- Expected: explanation should be in Japanese, matching the language setting
Additional Context
- Related closed issues: #23592 (closed as dup of #4866), #4866 (static UI i18n)
- The explanation model is already generating natural language — adding a language instruction to its prompt should be a minimal change
This issue has 2 comments on GitHub. Read the full discussion on GitHub ↗