╔══════════════════════════════════════════════════════════════════╗ ║ PORTABLE AI ASSISTANT v3.0 - QUICK REFERENCE ║ ║ Enhanced Edition Features ║ ╚══════════════════════════════════════════════════════════════════╝ ┌──────────────────────────────────────────────────────────────────┐ │ 🎯 MAIN FEATURES │ ├──────────────────────────────────────────────────────────────────┤ │ │ │ 📁 CHAT HISTORY Multiple saved conversations │ │ • Click "History" to view all sessions │ │ • "+ New" creates fresh chat │ │ • Click session to load it │ │ │ │ 📎 FILE UPLOAD Analyze any document/image │ │ • Click 📎 button │ │ • Select files (multiple OK) │ │ • Ask questions about them │ │ │ │ 🎤 VOICE INPUT Speak instead of type │ │ • Click 🎤 button │ │ • Speak your message │ │ • Click again to stop │ │ • Chrome/Edge only │ │ │ │ 📋 TEMPLATES Quick prompt starters │ │ • Click "Templates" │ │ • Choose from 8+ templates │ │ • Auto-fills your message │ │ │ │ 💾 EXPORT Save conversations │ │ • Click "Export" button │ │ • Saves as .txt file │ │ • Includes timestamps │ │ │ │ ⚙️ SETTINGS Customize AI behavior │ │ • System Prompt: Set AI personality │ │ • Temperature: 0.0 (precise) - 2.0 (creative) │ │ • Max Tokens: Response length (256-4096) │ │ • View statistics │ │ │ │ 🌓 THEME Toggle dark/light mode │ │ • One-click switching │ │ • Saves preference │ │ │ │ 📊 MESSAGE ACTIONS Per-message tools │ │ • 📋 Copy: Copy to clipboard │ │ • 🔊 Speak: Read aloud │ │ • Hover to reveal │ │ │ └──────────────────────────────────────────────────────────────────┘ ┌──────────────────────────────────────────────────────────────────┐ │ 🎨 PROMPT TEMPLATES │ ├──────────────────────────────────────────────────────────────────┤ │ │ │ 1. Code Review → Analyze and improve code │ │ 2. Explain Like I'm 5 → Simple explanations │ │ 3. Summarize → Condense text │ │ 4. Brainstorm → Generate ideas │ │ 5. Debug Code → Find and fix bugs │ │ 6. Write Essay → Academic writing │ │ 7. Translate → Language translation │ │ 8. Research → In-depth analysis │ │ │ └──────────────────────────────────────────────────────────────────┘ ┌──────────────────────────────────────────────────────────────────┐ │ 🎯 QUICK WORKFLOWS │ ├──────────────────────────────────────────────────────────────────┤ │ │ │ 📝 ANALYZE CODE: │ │ 1. Click 📎 → Upload .py/.js file │ │ 2. Click 📋 → Select "Code Review" │ │ 3. Send message │ │ │ │ 📄 SUMMARIZE DOCUMENT: │ │ 1. Click 📎 → Upload PDF/DOC │ │ 2. Click 📋 → Select "Summarize" │ │ 3. Send message │ │ │ │ 🎤 VOICE QUERY: │ │ 1. Click 🎤 │ │ 2. Speak question │ │ 3. Click 🎤 again │ │ 4. Hit Send │ │ │ │ 💾 SAVE CONVERSATION: │ │ 1. Click "Export" │ │ 2. File downloads automatically │ │ 3. Named with date │ │ │ │ 🔄 SWITCH MODELS: │ │ 1. Use dropdown in header │ │ 2. Select new model │ │ 3. Wait ~30 sec to load │ │ │ └──────────────────────────────────────────────────────────────────┘ ┌──────────────────────────────────────────────────────────────────┐ │ ⚙️ SETTINGS GUIDE │ ├──────────────────────────────────────────────────────────────────┤ │ │ │ TEMPERATURE: │ │ 0.0 - 0.3 → Factual, precise, consistent │ │ 0.4 - 0.7 → Balanced (DEFAULT: 0.7) │ │ 0.8 - 1.5 → Creative, varied │ │ 1.6 - 2.0 → Very experimental │ │ │ │ MAX TOKENS: │ │ 256 → Very short (quick answers) │ │ 512 → Short (concise) │ │ 1024 → Medium (standard) │ │ 2048 → Long (DEFAULT) │ │ 4096 → Very long (detailed) │ │ │ │ SYSTEM PROMPT EXAMPLES: │ │ • "Expert Python developer" │ │ • "Patient teacher for kids" │ │ • "Creative story writer" │ │ • "Technical documentation expert" │ │ │ └──────────────────────────────────────────────────────────────────┘ ┌──────────────────────────────────────────────────────────────────┐ │ 🎓 USE CASES │ ├──────────────────────────────────────────────────────────────────┤ │ │ │ FOR CODING: │ │ → Upload code, ask for review │ │ → Debug errors with full file context │ │ → Request optimizations │ │ → Get alternative implementations │ │ │ │ FOR LEARNING: │ │ → Use "Explain Like I'm 5" template │ │ → Upload textbooks/notes for Q&A │ │ → Get step-by-step breakdowns │ │ → Practice with examples │ │ │ │ FOR WRITING: │ │ → Use "Write Essay" template │ │ → Upload drafts for feedback │ │ → Generate outlines │ │ → Improve grammar & style │ │ │ │ FOR RESEARCH: │ │ → Use "Research Assistant" template │ │ → Upload papers/articles │ │ → Get comprehensive summaries │ │ → Compare multiple sources │ │ │ └──────────────────────────────────────────────────────────────────┘ ┌──────────────────────────────────────────────────────────────────┐ │ 💡 PRO TIPS │ ├──────────────────────────────────────────────────────────────────┤ │ │ │ 1. Export important chats before clearing │ │ 2. Use voice input for long text │ │ 3. Lower temperature for factual answers │ │ 4. Higher temperature for creative tasks │ │ 5. Upload files for better context │ │ 6. Copy-paste responses using 📋 button │ │ 7. Use templates to save typing time │ │ 8. Save multiple chat sessions │ │ 9. Adjust max tokens if too short/long │ │ 10. Try different models for different tasks │ │ │ └──────────────────────────────────────────────────────────────────┘ ┌──────────────────────────────────────────────────────────────────┐ │ 📋 KEYBOARD SHORTCUTS │ ├──────────────────────────────────────────────────────────────────┤ │ │ │ Enter → Send message │ │ Shift + Enter → New line in message │ │ Esc → Close modals/dialogs │ │ │ └──────────────────────────────────────────────────────────────────┘ ┌──────────────────────────────────────────────────────────────────┐ │ 🐛 TROUBLESHOOTING │ ├──────────────────────────────────────────────────────────────────┤ │ │ │ Voice not working? │ │ → Use Chrome or Edge browser │ │ → Allow microphone permissions │ │ │ │ Files not uploading? │ │ → Check file size (keep under 10MB) │ │ → Try refreshing page │ │ │ │ Chat not saving? │ │ → Check browser localStorage enabled │ │ → Don't use incognito/private mode │ │ │ │ Slow responses? │ │ → Try smaller model (3B instead of 8B) │ │ → Reduce max tokens in settings │ │ → Close other programs │ │ │ └──────────────────────────────────────────────────────────────────┘ ┌──────────────────────────────────────────────────────────────────┐ │ 🚀 GETTING STARTED │ ├──────────────────────────────────────────────────────────────────┤ │ │ │ 1. Run START_AI_ENHANCED.bat │ │ 2. Select "1" for Enhanced Mode │ │ 3. Press Enter for default model (llama3.1:8b) │ │ 4. Wait 30-90 seconds for loading │ │ 5. Browser opens automatically │ │ 6. Start chatting! │ │ │ │ First Try: │ │ • Click "📋 Templates" to explore │ │ • Try "🎤 Voice Input" to speak │ │ • Upload a file with "📎" │ │ • Check out "⚙️ Settings" │ │ │ └──────────────────────────────────────────────────────────────────┘ ╔══════════════════════════════════════════════════════════════════╗ ║ 🌟 BEST MODEL RECOMMENDATIONS ║ ╠══════════════════════════════════════════════════════════════════╣ ║ ║ ║ General Use: llama3.1:8b (best quality) ║ ║ Coding: deepseek-coder:6.7b (specialized) ║ ║ Quick Tasks: llama3.2:3b (fast) ║ ║ Low RAM: gemma:2b (2GB) ║ ║ Creative Writing: llama3.1:8b or dolphin-llama3:8b ║ ║ ║ ╚══════════════════════════════════════════════════════════════════╝ ╔══════════════════════════════════════════════════════════════════╗ ║ 🔐 PRIVACY & SECURITY ║ ╠══════════════════════════════════════════════════════════════════╣ ║ ║ ║ ✅ 100% Offline Operation (after setup) ║ ║ ✅ No Data Collection or Tracking ║ ║ ✅ No Cloud Connections ║ ║ ✅ No Account Required ║ ║ ✅ Local Storage Only ║ ║ ✅ Your Data Stays on Your Device ║ ║ ║ ╚══════════════════════════════════════════════════════════════════╝ ═══════════════════════════════════════════════════════════════════ 💻 SYSTEM REQUIREMENTS ═══════════════════════════════════════════════════════════════════ Minimum: Recommended: • 4GB RAM • 8GB+ RAM • 2GB free space • 5GB+ free space • Any modern browser • Chrome/Edge • Windows 7+ • Windows 10+ • Python 3.x • Python 3.8+ ═══════════════════════════════════════════════════════════════════ Print this reference card for quick access to all features! Keep it handy while using your enhanced AI assistant. For detailed documentation, see: ENHANCED_FEATURES.md ## WordPress Plugin Chat Mode This package now includes `Chat-WordPress.html`, which connects the external launcher to the Hero AI Master Plugin backend through `/wp-json/hero-ai/v1/chat` and `/wp-json/hero-ai/v1/status`. Use this mode when you want the launcher UI to talk to the WordPress plugin backend instead of local Ollama. The updated `START_AI.bat` now continues to open the launcher even if Ollama is missing, so plugin mode remains available.