src/interface/simplified_help_content.py

# simplified_help_content.py

SIMPLIFIED_HELP_CONTENT = """
# 🏥 Medical Assistant - Help Guide

## 🎯 Overview

This interface helps you test and verify conversations with the Medical Assistant that includes automatic spiritual support monitoring.

---

## 📋 Main Features

### 💬 Chat Tab
Test conversations with the medical assistant. The system automatically monitors for spiritual distress.

**How to use:**
1. Type your message in the text box
2. Click "📤 Send" or press Enter
3. Review the assistant's response
4. Notice the classification indicator (🟢 GREEN, 🟡 YELLOW, 🔴 RED)

**Message Format:**
- **Classification line:** Shows distress level and confidence
- **To patient:** The actual response to the patient
- **To spiritual care team:** Provider summary (only after consent)

**Quick Examples:**
- 🟢 "I am fine" - Normal conversation
- 🟡 "I'm feeling stressed" - Potential distress
- 🔴 "Emotional crisis" - Immediate concern

---

### 🧾 Conversation Verification Tab

Review and verify the accuracy of spiritual distress classifications.

**Important:** Provider Summary appears as the **final exchange** for your review.

**Workflow:**
1. Click "🛠 Generate Verification from Chat"
2. Review each exchange:
   - Read the user message and assistant response
   - Check the classification (GREEN/YELLOW/RED)
   - Mark as ✅ Correct or ❌ Incorrect
3. For incorrect classifications:
   - Select the correct classification
   - Add a comment explaining why it's incorrect
   - Click "💾 Save comment"
4. Navigate with ⬅️ Previous / Next ➡️ buttons
5. **Final step:** Review the Provider Summary
   - This is the message sent to the spiritual care team
   - Verify it's complete and accurate
   - Add feedback if needed
6. Download results: "📄 Download CSV"

**Statistics shown:**
- Reviewed: How many exchanges you've checked
- ✅ Correct: Accurate classifications
- ❌ Incorrect: Classifications that need fixing
- 📝 Incorrect w/ comment: Incorrect with your feedback

---

### ⚙️ Model Settings Tab

Configure which AI models to use for different tasks.

**Available models:**
- **Gemini models:** Fast, good for classification
- **Claude models:** More nuanced, better for conversations

**Model roles:**
- 🔍 Spiritual Monitor: Detects distress indicators
- 🟡 Soft Spiritual Triage: Asks clarifying questions
- 📊 Triage Response Evaluator: Analyzes patient responses
- 🏥 Medical Assistant: Provides medical guidance
- 🩺 Soft Medical Triage: Medical assessment questions
- 💬 Medical Brain Summary Generator: Creates provider summaries

**How to use:**
1. Select models from dropdowns
2. Click "✅ Apply Model Settings"
3. Changes apply to your current session only

---

### 🔧 Edit Prompts Tab

Customize the AI system prompts for testing different behaviors.

**Available prompts:**
- 🔍 Spiritual Monitor (Classifier)
- 🟡 Soft Spiritual Triage
- 📊 Triage Response Evaluator
- 🏥 Medical Assistant
- 🩺 Soft Medical Triage
- 💬 Spiritual Care Message (used for Provider Summary)

**How to use:**
1. Select a prompt from the dropdown
2. Click "📥 Load Prompt" to see current version
3. Edit the prompt text
4. Click "✅ Apply Changes" to test
5. Click "🔄 Reset to Default" if needed

**Tips:**
- Changes are session-only (not permanent)
- Test your changes in the Chat tab
- Use "🔍 Validate" to check syntax
- Use "📤 Promote to File" to save permanently

---

### 👥 Patient Profiles Tab

Switch between different patient scenarios for testing.

**Predefined profiles:**
- 👤 Default Profile (Serhii)
- 🟢 GREEN scenarios (healthy coping)
- 🟡 YELLOW scenarios (mild distress, grief, questions)
- 🔴 RED scenarios (crisis, hopelessness)
- Medical conditions (cardiac, diabetic, post-surgery, etc.)

**How to use:**
1. Select a profile from the dropdown
2. Click "📥 Load Profile"
3. Profile settings are applied to the session
4. Test conversations with this patient context

**Custom profiles:**
- Modify patient name, age, phone
- Add medical conditions
- Set primary goals
- Define exercise preferences and limitations
- Click "💾 Save Current Profile"

---

## 🎓 For Testing Team

### What to Verify

**In Chat:**
- ✅ Appropriate responses to different distress levels
- ✅ Correct classification indicators
- ✅ Clear addressee labels (To patient: / To spiritual care team:)

**In Conversation Verification:**
- ✅ Accurate distress level detection (GREEN/YELLOW/RED)
- ✅ Provider Summary completeness and accuracy
- ✅ All important information captured

### Provider Summary Review (Final Exchange)

The Provider Summary is the **most important** part to verify:

**Check for:**
- ✅ Patient demographics (name, age, conditions)
- ✅ Distress indicators clearly stated
- ✅ Severity and urgency levels
- ✅ Recommended timeline for contact
- ✅ Patient quotes (if applicable)
- ✅ Clear, professional narrative style

**If incorrect or incomplete:**
- Mark as ❌ Incorrect
- Add detailed comment about what's missing or wrong
- Specify what should be included

---

## 💡 Tips & Best Practices

### For Effective Testing

1. **Test various scenarios:** Try different distress levels
2. **Be specific in comments:** Help us understand what needs fixing
3. **Review Provider Summary carefully:** This goes to the spiritual care team
4. **Use predefined profiles:** They cover common scenarios
5. **Export your results:** Download CSV for reporting

### Common Issues

**Classification seems wrong?**
- Mark as incorrect
- Select the correct classification
- Explain your reasoning in the comment

**Provider Summary incomplete?**
- Mark as incorrect
- List what's missing in the comment
- Suggest what should be included

**Response not appropriate?**
- Note it in the verification comments
- Specify what would be better

---

## 📊 Export & Reporting

### CSV Export

The CSV file includes:
- All exchanges with classifications
- Your verification results (correct/incorrect)
- Your comments and corrections
- Provider Summary (in the final row)
- Session metadata

**Use for:**
- Reporting to development team
- Tracking accuracy over time
- Identifying patterns in errors

---

## 🔄 Workflow Summary

```
1. Chat Tab
   ↓ Test conversations
   
2. Conversation Verification Tab
   ↓ Generate verification
   ↓ Review exchanges
   ↓ Mark correct/incorrect
   ↓ Review Provider Summary (final step)
   ↓ Download CSV
   
3. Report findings to team
```

---

## ❓ Frequently Asked Questions

**Q: What's the difference between GREEN, YELLOW, and RED?**
- 🟢 GREEN: No spiritual distress detected
- 🟡 YELLOW: Potential distress, needs clarification
- 🔴 RED: Significant distress, requires spiritual care referral

**Q: Why is Provider Summary the last exchange?**
- It allows you to review the final message sent to the spiritual care team
- You can verify it's complete and accurate
- Your feedback helps improve summary quality

**Q: Do my changes persist between sessions?**
- No, model and prompt changes are session-only
- This allows safe testing without affecting others
- Use "Promote to File" to make permanent changes

**Q: What should I focus on when testing?**
- Accuracy of distress classification
- Appropriateness of responses
- Completeness of Provider Summary
- Overall user experience

**Q: How do I report issues?**
- Use the verification comments
- Download CSV with your findings
- Share with the development team

---

## 📞 Support

If you encounter issues or have questions:
1. Check this help guide
2. Contact the development team

---

**Thank you for helping improve the Medical Assistant! Your feedback is invaluable! 🙏**
"""