Tool / Prompt / Output
AI / GenAI DocumentationAI Tool
| Property | Value |
|---|---|
| Model | Google Gemini 2.5 Flash Lite (gemini-2.5-flash-lite) |
| SDK | google-generativeai Python SDK |
| Capabilities | Text generation, multimodal image analysis, contextual translation |
| Output Format | Structured JSON with empathetic_response, doctor_summary, comfort_measures |
| Safety | JSON validation, response truncation, keyword fallback |
Prompt: Symptom Classification
Pregnancy assistant, week {week}. Respond in {language} with JSON only. Keep responses brief: { "empathetic_response": "Brief supportive message (max 50 words)...", "comfort_measures": "Short list (max 20 words)..." }
Prompt: Image Analysis
Analyze this image in the context of pregnancy week {week}. The user asks: {question}. Respond with JSON: { "empathetic_response": "...", "comfort_measures": "..." }
Prompt: Contextual Translation
Translate the following pregnancy-related message to Korean. Context: pregnancy week {week}. Message: {message}
Sample Output
Input (Vietnamese): 'Toi bi dau dau nhieu hom nay' Output: { "empathetic_response": "I understand headaches can be really uncomfortable, especially during pregnancy...", "doctor_summary": "임신 32주 산모, 오늘 심한 두통 호소...", "comfort_measures": "Rest, hydration, dim lighting" }