Language Settings Skill

Manage user language preferences for the health management system. Supports Chinese (zh-CN) and English (en).

Core Flow

User Input -> Parse Operation -> [set] Update preferred language -> Save
                             -> [get] Read current settings -> Display
                             -> [auto] Enable/disable auto-detection -> Save

Step 1: Parse Operation Type

Step 2: Set Language (set)

Supported Languages

Examples

/language set en
/language set zh-CN
/language switch to English

Step 3: Get Language (get)

Display current language settings:

/language get
/language show

Response format:

• Preferred language
• Auto-detection status
• Fallback language

Step 4: Auto-Detection (auto)

Enable or disable automatic language detection from user input.

/language auto on
/language auto off

Data Structure

The language skill reads and updates data/user-settings.json:

{
  "language": {
    "preferred": "en",
    "detected": "en",
    "detected_at": "2026-02-03T10:00:00Z",
    "confidence": 0.95
  },
  "onboarding": {
    "completed": false,
    "language_set": false,
    "steps_completed": []
  },
  "conversation": {
    "auto_detect": true,
    "fallback_language": "en"
  },
  "metadata": {
    "created_at": "2026-02-03T10:00:00Z",
    "last_updated": "2026-02-03T10:00:00Z",
    "version": "1.0.0"
  }
}

Execution Instructions

1. Parse operation type
2. [set] Validate language code -> Update user-settings.json -> Confirm
3. [get] Read user-settings.json -> Format display
4. [auto] Toggle auto_detect flag -> Save

Example Interactions

Set Language

User: /language set en

Response:
Language preference updated to English.
Your preferred language: en

Get Language

User: /language get

Response:
Current language settings:
- Preferred language: English (en)
- Auto-detection: enabled
- Fallback language: English (en)

Disable Auto-Detection

User: /language auto off

Response:
Auto language detection disabled.
The system will always use your preferred language: en

Language Detection Logic

When auto_detect is enabled:

1. Analyze user input for Chinese character ratio
2. If Chinese characters > 30%, detect as zh-CN
3. Otherwise, detect as en
4. Store detection result in detected field
5. Use detected language for current response