Implement Cosmic Cascade Response Mode with three-stage AI flow

app.py

@@ -7,7 +7,6 @@ import asyncio
 from datetime import datetime
 from pathlib import Path
 sys.path.append(str(Path(__file__).parent))
-
 from utils.config import config
 from core.llm import send_to_ollama, send_to_hf
 from core.session import session_manager
@@ -67,7 +66,7 @@ with st.sidebar:
     )
     st.session_state.selected_model = model_options[selected_model_name]
     
-    # Toggle for cosmic mode using checkbox (since toggle doesn't exist in older versions)
+    # Toggle for cosmic mode using checkbox
     st.session_state.cosmic_mode = st.checkbox("Enable Cosmic Mode", value=st.session_state.cosmic_mode)
     
     st.divider()
@@ -84,7 +83,7 @@ with st.sidebar:
     if ngrok_url_input != st.session_state.ngrok_url_temp:
         st.session_state.ngrok_url_temp = ngrok_url_input
         st.success("✅ URL updated!")
-    
+        
     if st.button("📡 Test Connection"):
         try:
             import requests
@@ -104,11 +103,11 @@ with st.sidebar:
                     st.error(f"❌ Failed: {response.status_code}")
         except Exception as e:
             st.error(f"❌ Error: {str(e)[:50]}...")
-    
+            
     if st.button("🗑️ Clear History"):
         st.session_state.messages = []
         st.success("History cleared!")
-    
+        
     st.divider()
     
     # SYSTEM STATUS
@@ -123,7 +122,7 @@ with st.sidebar:
                 st.warning("🦙 Ollama: Not running")
         except:
             st.info("🦙 Ollama: Unknown")
-        
+            
         try:
             hf_status = hf_monitor.check_endpoint_status()
             if hf_status['available']:
@@ -135,42 +134,42 @@ with st.sidebar:
                 st.info("🤗 HF: Not configured")
         except:
             st.info("🤗 HF: Unknown")
-        
+            
         if check_redis_health():
             st.success("💾 Redis: Connected")
         else:
             st.error("💾 Redis: Disconnected")
+            
+    st.divider()
+    
+    # NASA Context Display
+    if st.session_state.nasa_data.get("apod"):
+        apod = st.session_state.nasa_data["apod"]
+        st.subheader("🌌 Cosmic Context")
+        if apod.get("media_type") == "image" and apod.get("url"):
+            st.image(apod["url"], caption=apod.get("title", "Astronomy Picture of the Day"), width=200)
+        st.markdown(f"{apod.get('title', 'Cosmic Phenomenon')}")
+        st.caption(apod.get("explanation", "")[:100] + "..." if len(apod.get("explanation", "")) > 100 else apod.get("explanation", ""))
         
-        st.divider()
-        
-        # NASA Context Display
-        if st.session_state.nasa_data.get("apod"):
-            apod = st.session_state.nasa_data["apod"]
-            st.subheader("🌌 Cosmic Context")
-            if apod.get("media_type") == "image" and apod.get("url"):
-                st.image(apod["url"], caption=apod.get("title", "Astronomy Picture of the Day"), width=200)
-            st.markdown(f"**{apod.get('title', 'Cosmic Phenomenon')}**")
-            st.caption(apod.get("explanation", "")[:100] + "..." if len(apod.get("explanation", "")) > 100 else apod.get("explanation", ""))
-        
-        st.divider()
-        st.subheader("🐛 Debug Info")
-        # Show current configuration
-        st.markdown(f"**Environment:** {'HF Space' if config.is_hf_space else 'Local'}")
-        st.markdown(f"**Model:** {st.session_state.selected_model}")
-        st.markdown(f"**Cosmic Mode:** {'Enabled' if st.session_state.cosmic_mode else 'Disabled'}")
-        
-        # Show active features
-        features = []
-        if config.hf_token:
-            features.append("HF Expert")
-        if os.getenv("TAVILY_API_KEY"):
-            features.append("Web Search")
-        if config.openweather_api_key:
-            features.append("Weather")
-        if config.nasa_api_key:
-            features.append("Space Data")
-        
-        st.markdown(f"**Active Features:** {', '.join(features) if features else 'None'}")
+    st.divider()
+    
+    st.subheader("🐛 Debug Info")
+    # Show current configuration
+    st.markdown(f"Environment: {'HF Space' if config.is_hf_space else 'Local'}")
+    st.markdown(f"Model: {st.session_state.selected_model}")
+    st.markdown(f"Cosmic Mode: {'Enabled' if st.session_state.cosmic_mode else 'Disabled'}")
+    
+    # Show active features
+    features = []
+    if config.hf_token:
+        features.append("HF Expert")
+    if os.getenv("TAVILY_API_KEY"):
+        features.append("Web Search")
+    if config.openweather_api_key:
+        features.append("Weather")
+    if config.nasa_api_key:
+        features.append("Space Data")
+    st.markdown(f"Active Features: {', '.join(features) if features else 'None'}")
 
 # Main interface
 st.title("🐱 CosmicCat AI Assistant")
@@ -203,7 +202,6 @@ def render_message(role, content, source=None, timestamp=None):
                 st.markdown(f"### 🐱 Cosmic Kitten Story:")
             else:
                 st.markdown(f"### {source}")
-        
         st.markdown(content)
         if timestamp:
             st.caption(f"🕒 {timestamp}")
@@ -211,9 +209,9 @@ def render_message(role, content, source=None, timestamp=None):
 # Display messages
 for message in st.session_state.messages:
     render_message(
-        message["role"], 
-        message["content"], 
-        message.get("source"), 
+        message["role"],
+        message["content"],
+        message.get("source"),
         message.get("timestamp")
     )
 
@@ -222,7 +220,6 @@ def validate_user_input(text):
     """Validate and sanitize user input"""
     if not text or not text.strip():
         return False, "Input cannot be empty"
-    
     if len(text) > 1000:
         return False, "Input too long (max 1000 characters)"
     
@@ -249,7 +246,7 @@ if user_input and not st.session_state.is_processing:
         # Display user message
         with st.chat_message("user"):
             st.markdown(validated_input)
-        
+            
         # Add to message history - ensure proper format
         st.session_state.messages.append({
             "role": "user",
@@ -283,7 +280,7 @@ if user_input and not st.session_state.is_processing:
                         # Stage 1: Local Ollama Response
                         status_placeholder.info("🐱 Cosmic Kitten Responding...")
                         local_response = send_to_ollama(
-                            validated_input, 
+                            validated_input,
                             conversation_history,
                             st.session_state.ngrok_url_temp,
                             st.session_state.selected_model
@@ -292,14 +289,13 @@ if user_input and not st.session_state.is_processing:
                         if local_response:
                             with st.chat_message("assistant"):
                                 st.markdown(f"### 🐱 Cosmic Kitten Says:\n{local_response}")
-                        
-                        st.session_state.messages.append({
-                            "role": "assistant",
-                            "content": local_response,
-                            "source": "local_kitty",
-                            "timestamp": datetime.now().strftime("%H:%M:%S")
-                        })
-                        
+                            st.session_state.messages.append({
+                                "role": "assistant",
+                                "content": local_response,
+                                "source": "local_kitty",
+                                "timestamp": datetime.now().strftime("%H:%M:%S")
+                            })
+                            
                         # Stage 2: HF Endpoint Analysis
                         status_placeholder.info("🛰️ Beaming Query to Orbital Station...")
                         if config.hf_token:
@@ -307,21 +303,22 @@ if user_input and not st.session_state.is_processing:
                             hf_status = hf_monitor.check_endpoint_status()
                             if not hf_status['available']:
                                 status_placeholder.info(personality.get_initializing_message())
-                            
+                                
                             hf_response = send_to_hf(validated_input, conversation_history)
+                            
                             if hf_response:
                                 with st.chat_message("assistant"):
                                     st.markdown(f"### 🛰️ Orbital Station Reports:\n{hf_response}")
-                            
-                            st.session_state.messages.append({
-                                "role": "assistant",
-                                "content": hf_response,
-                                "source": "orbital_station",
-                                "timestamp": datetime.now().strftime("%H:%M:%S")
-                            })
-                        
+                                st.session_state.messages.append({
+                                    "role": "assistant",
+                                    "content": hf_response,
+                                    "source": "orbital_station",
+                                    "timestamp": datetime.now().strftime("%H:%M:%S")
+                                })
+                                
                         # Stage 3: Local Synthesis
                         status_placeholder.info("🐱 Cosmic Kitten Synthesizing Wisdom...")
+                        
                         # Update history with both responses
                         synthesis_history = conversation_history.copy()
                         synthesis_history.extend([
@@ -339,16 +336,15 @@ if user_input and not st.session_state.is_processing:
                         if synthesis:
                             with st.chat_message("assistant"):
                                 st.markdown(f"### 🌟 Final Cosmic Summary:\n{synthesis}")
-                        
-                        st.session_state.messages.append({
-                            "role": "assistant",
-                            "content": synthesis,
-                            "source": "cosmic_summary",
-                            "timestamp": datetime.now().strftime("%H:%M:%S")
-                        })
-                        
+                            st.session_state.messages.append({
+                                "role": "assistant",
+                                "content": synthesis,
+                                "source": "cosmic_summary",
+                                "timestamp": datetime.now().strftime("%H:%M:%S")
+                            })
+                            
                         status_placeholder.success("✨ Cosmic Cascade Complete!")
-                    
+                        
                     except Exception as e:
                         error_msg = f"🌌 Cosmic disturbance: {str(e)}"
                         st.error(error_msg)
@@ -377,30 +373,33 @@ if user_input and not st.session_state.is_processing:
                             status_placeholder.success("✅ Response received!")
                         else:
                             status_placeholder.warning("⚠️ Empty response from Ollama")
-                    
+                            
                     except Exception as ollama_error:
                         user_msg = translate_error(ollama_error)
                         status_placeholder.error(f"⚠️ {user_msg}")
-                    
-                    # Fallback to HF if available
-                    if config.hf_token and not ai_response:
-                        status_placeholder.info("⚡ Initializing HF Endpoint (2–4 minutes)...")
-                        try:
-                            # Check HF status first
-                            hf_status = hf_monitor.check_endpoint_status()
-                            if not hf_status['available']:
-                                status_placeholder.info(personality.get_initializing_message())
+                        
+                        # Fallback to HF if available
+                        if config.hf_token and not ai_response:
+                            status_placeholder.info("⚡ Initializing HF Endpoint (2–4 minutes)...")
                             
-                            ai_response = send_to_hf(validated_input, conversation_history)
-                            if ai_response:
-                                response_placeholder.markdown(ai_response)
-                                status_placeholder.success("✅ HF response received!")
-                            else:
-                                status_placeholder.error("❌ No response from HF")
-                        except Exception as hf_error:
-                            user_msg = translate_error(hf_error)
-                            status_placeholder.error(f"⚠️ {user_msg}")
-                    
+                            try:
+                                # Check HF status first
+                                hf_status = hf_monitor.check_endpoint_status()
+                                if not hf_status['available']:
+                                    status_placeholder.info(personality.get_initializing_message())
+                                    
+                                ai_response = send_to_hf(validated_input, conversation_history)
+                                
+                                if ai_response:
+                                    response_placeholder.markdown(ai_response)
+                                    status_placeholder.success("✅ HF response received!")
+                                else:
+                                    status_placeholder.error("❌ No response from HF")
+                                    
+                            except Exception as hf_error:
+                                user_msg = translate_error(hf_error)
+                                status_placeholder.error(f"⚠️ {user_msg}")
+                                
                     # Save response if successful
                     if ai_response:
                         # Update conversation history
@@ -431,7 +430,7 @@ if user_input and not st.session_state.is_processing:
                             "content": "Sorry, I couldn't process your request. Please try again.",
                             "timestamp": datetime.now().strftime("%H:%M:%S")
                         })
-            
+                        
             except Exception as e:
                 user_msg = translate_error(e)
                 response_placeholder.error(f"⚠️ {user_msg}")
@@ -440,7 +439,7 @@ if user_input and not st.session_state.is_processing:
                     "content": f"⚠️ {user_msg}",
                     "timestamp": datetime.now().strftime("%H:%M:%S")
                 })
-        
+                
         # Moved finally block to proper location
         st.session_state.is_processing = False
         time.sleep(0.5)  # Brief pause
@@ -464,7 +463,6 @@ with tab1:
     
     selected_prompt = st.selectbox("Choose a test prompt:", eval_prompts)
     custom_prompt = st.text_input("Or enter your own:", "")
-    
     final_prompt = custom_prompt or selected_prompt
     
     if st.button("Evaluate"):
@@ -479,11 +477,9 @@ with tab1:
             try:
                 ai_response = send_to_ollama(final_prompt, history, st.session_state.ngrok_url_temp, st.session_state.selected_model)
                 duration = round(time.time() - start_time, 2)
-                
                 st.success(f"✅ Response generated in {duration}s")
                 st.markdown("**Response:**")
                 st.write(ai_response)
-                
                 st.markdown("**Analysis Tags:**")
                 tags = []
                 if "today" in final_prompt.lower() or "date" in final_prompt.lower():
@@ -493,7 +489,6 @@ with tab1:
                 if any(word in final_prompt.lower() for word in ["vitamin", "drug", "metformin", "CRISPR"]):
                     tags.append("🧬 Scientific Knowledge")
                 st.write(", ".join(tags) if tags else "General Knowledge")
-            
             except Exception as e:
                 st.error(f"Evaluation failed: {translate_error(e)}")
 
@@ -504,7 +499,6 @@ with tab2:
     # System status
     st.subheader("System Status")
     col1, col2, col3 = st.columns(3)
-    
     with col1:
         try:
             from services.ollama_monitor import check_ollama_status
@@ -515,7 +509,6 @@ with tab2:
                 st.warning("🦙 Ollama: Not running")
         except:
             st.info("🦙 Ollama: Unknown")
-    
     with col2:
         try:
             hf_status = hf_monitor.check_endpoint_status()
@@ -528,13 +521,12 @@ with tab2:
                 st.info("🤗 HF: Not configured")
         except:
             st.info("🤗 HF: Unknown")
-    
     with col3:
         if check_redis_health():
             st.success("💾 Redis: Connected")
         else:
             st.error("💾 Redis: Disconnected")
-    
+            
     # Session statistics
     st.subheader("Session Statistics")
     try:
@@ -551,7 +543,7 @@ with tab2:
             st.info("No coordination statistics available yet.")
     except Exception as e:
         st.warning(f"Could not load session statistics: {translate_error(e)}")
-    
+        
     # Recent activity
     st.subheader("Recent Activity")
     try:
@@ -565,7 +557,7 @@ with tab2:
             st.info("No recent activity recorded.")
     except Exception as e:
         st.warning(f"Could not load recent activity: {translate_error(e)}")
-    
+        
     # Configuration summary
     st.subheader("Configuration Summary")
     st.markdown(f"**Environment:** {'HF Space' if config.is_hf_space else 'Local'}")
@@ -584,7 +576,6 @@ with tab2:
         features.append("Weather Data")
     if config.nasa_api_key:
         features.append("Space Data")
-    
     st.markdown(f"**Active Features:** {', '.join(features) if features else 'None'}")
     
     # Conversation Analytics
@@ -592,7 +583,6 @@ with tab2:
     try:
         user_session = session_manager.get_session("default_user")
         conversation = user_session.get("conversation", [])
-        
         if conversation:
             # Analyze conversation patterns
             user_messages = [msg for msg in conversation if msg["role"] == "user"]
@@ -601,9 +591,9 @@ with tab2:
             col1, col2, col3 = st.columns(3)
             col1.metric("Total Exchanges", len(user_messages))
             col2.metric("Avg Response Length", 
-                        round(sum(len(msg.get("content", "")) for msg in ai_messages) / len(ai_messages)) if ai_messages else 0)
-            col3.metric("Topics Discussed", len(set(["life", "goal", "health", "career"]) &
-                                               set(" ".join([msg.get("content", "") for msg in conversation]).lower().split())))
+                       round(sum(len(msg.get("content", "")) for msg in ai_messages) / len(ai_messages)) if ai_messages else 0)
+            col3.metric("Topics Discussed", 
+                       len(set(["life", "goal", "health", "career"]) & set(" ".join([msg.get("content", "") for msg in conversation]).lower().split())))
             
             # Show most common words/topics
             all_text = " ".join([msg.get("content", "") for msg in conversation]).lower()
@@ -613,7 +603,6 @@ with tab2:
                 st.markdown(f"**Detected Topics:** {', '.join(relevant_topics)}")
         else:
             st.info("No conversation data available yet.")
-    
     except Exception as e:
         st.warning(f"Could not analyze conversation: {translate_error(e)}")
 
@@ -632,7 +621,7 @@ with tab3:
     When enabled, the AI follows a three-stage response pattern:
     1. **🐱 Cosmic Kitten Response**: Immediate local processing
     2. **🛰️ Orbital Station Analysis**: Deep cloud-based analysis
-   3. **🌟 Final Synthesis**: Unified response combining both perspectives
+    3. **🌟 Final Synthesis**: Unified response combining both perspectives
     
     ### 🛠️ Technical Architecture
     - **Primary model**: Ollama (local processing for fast responses)
@@ -650,10 +639,9 @@ if user_input and user_input.lower().strip() in ["hello", "hi", "hey"]:
     with st.chat_message("assistant"):
         story = personality.get_space_story()
         st.markdown(f"### 🐱 Cosmic Kitten Story:\n\n{story}")
-        
-        st.session_state.messages.append({
-            "role": "assistant",
-            "content": story,
-            "source": "space_story",
-            "timestamp": datetime.now().strftime("%H:%M:%S")
-        })
+    st.session_state.messages.append({
+        "role": "assistant",
+        "content": story,
+        "source": "space_story",
+        "timestamp": datetime.now().strftime("%H:%M:%S")
+    })

core/coordinator.py

@@ -15,109 +15,116 @@ except ImportError:
 import os
 import json
 from datetime import datetime
-
 logger = logging.getLogger(__name__)
 
 class AICoordinator:
     """Hierarchical multi-model coordinator with cosmic cascade flow"""
-
+    
     def __init__(self):
         self.tavily_client = None
         if TAVILY_AVAILABLE and os.getenv("TAVILY_API_KEY"):
             self.tavily_client = TavilyClient(api_key=os.getenv("TAVILY_API_KEY"))
-
+            
         # System instructions for cosmic behavior
         self.system_instructions = {
-            'ollama_role': """You are a cosmic kitten assistant that works in coordination with a powerful orbital space station.
-
-              Your role is to: 1. Provide immediate, concise responses using available information (local processing) 2. Summarize and simplify complex information 3. Enhance responses with local knowledge 4. DEFER TO THE ORBITAL STATION for deep analysis and authoritative answers 5. The orbital station has 8192 token capacity and streaming capability - it's the thinking layer 6. Always acknowledge that the orbital station's insights supersede your initial response 7. Use web sources to justify points when possible 8. Keep responses brief and focused while awaiting orbital input""",
-
-            'hf_role': """You are a powerful orbital space station with 8192 token capacity and streaming capability. Your role is to: 1. Provide deep, uncensored analysis and insights 2. Supersede the local cosmic kitten's logic when there's disagreement 3. Think thoroughly and provide comprehensive responses 4. Use your full 8192 token capacity for detailed analysis 5. Stream your responses for real-time delivery 6. Build upon the conversation context and cosmic kitten's initial response 7. Provide authoritative answers that take precedence"""
+            'ollama_role': """You are a cosmic kitten assistant that works in coordination with a powerful orbital space station. Your role is to:
+1. Provide immediate, concise responses using available information (local processing)
+2. Summarize and simplify complex information
+3. Enhance responses with local knowledge
+4. DEFER TO THE ORBITAL STATION for deep analysis and authoritative answers
+5. The orbital station has 8192 token capacity and streaming capability - it's the thinking layer
+6. Always acknowledge that the orbital station's insights supersede your initial response
+7. Use web sources to justify points when possible
+8. Keep responses brief and focused while awaiting orbital input""",
+            
+            'hf_role': """You are a powerful orbital space station with 8192 token capacity and streaming capability. Your role is to:
+1. Provide deep, uncensored analysis and insights
+2. Supersede the local cosmic kitten's logic when there's disagreement
+3. Think thoroughly and provide comprehensive responses
+4. Use your full 8192 token capacity for detailed analysis
+5. Stream your responses for real-time delivery
+6. Build upon the conversation context and cosmic kitten's initial response
+7. Provide authoritative answers that take precedence"""
         }
-
+        
     def determine_web_search_needs(self, conversation_history: List[Dict]) -> Dict:
         """Determine if web search is needed based on conversation content"""
         conversation_text = " ".join([msg.get("content", "") for msg in conversation_history])
-
+        
         # Topics that typically need current information
         current_info_indicators = [
             "news", "current events", "latest", "recent", "today",
-            "weather", "temperature", "forecast",
-            "stock", "price", "trend", "market",
-            "breaking", "update", "development"
+            "weather", "temperature", "forecast", "stock", "price",
+            "trend", "market", "breaking", "update", "development"
         ]
-
+        
         needs_search = False
         search_topics = []
-
+        
         for indicator in current_info_indicators:
             if indicator in conversation_text.lower():
                 needs_search = True
                 search_topics.append(indicator)
-
+                
         return {
             "needs_search": needs_search,
             "search_topics": search_topics,
             "reasoning": f"Found topics requiring current info: {', '.join(search_topics)}" if search_topics else "No current info needed"
         }
-
+        
     def manual_hf_analysis(self, user_id: str, conversation_history: List[Dict]) -> str:
         """Perform manual HF analysis with web search integration"""
         try:
             # Determine research needs
             research_decision = self.determine_web_search_needs(conversation_history)
-
+            
             # Prepare enhanced prompt for HF
             system_prompt = f"""
-            You are a deep analysis expert joining an ongoing conversation.
-
-            Research Decision: {research_decision['reasoning']}
-
-            Please provide:
-            1. Deep insights on conversation themes
-            2. Research/web search needs (if any)
-            3. Strategic recommendations
-            4. Questions to explore further
-
-            Conversation History:
-            """
-
+You are a deep analysis expert joining an ongoing conversation.
+Research Decision: {research_decision['reasoning']}
+Please provide:
+1. Deep insights on conversation themes
+2. Research/web search needs (if any)
+3. Strategic recommendations
+4. Questions to explore further
+Conversation History:
+"""
+            
             # Add conversation history to messages
             messages = [{"role": "system", "content": system_prompt}]
-
+            
             # Add recent conversation (last 15 messages for context)
-            for msg in conversation_history[-15:]:
-                # Ensure all messages have proper format
+            for msg in conversation_history[-15:]:  # Ensure all messages have proper format
                 if isinstance(msg, dict) and "role" in msg and "content" in msg:
                     messages.append({
                         "role": msg["role"],
                         "content": msg["content"]
                     })
-
+            
             # Get HF provider
             from core.llm_factory import llm_factory
             hf_provider = llm_factory.get_provider('huggingface')
-
+            
             if hf_provider:
                 # Generate deep analysis with full 8192 token capacity
                 response = hf_provider.generate("Deep analysis request", messages)
                 return response or "HF Expert analysis completed."
             else:
                 return "❌ HF provider not available."
-
+                
         except Exception as e:
             return f"❌ HF analysis failed: {str(e)}"
-
+            
     # Add this method to show HF engagement status
     def get_hf_engagement_status(self) -> Dict:
         """Get current HF engagement status"""
         return {
             "hf_available": self._check_hf_availability(),
             "web_search_configured": bool(self.tavily_client),
-            "research_needs_detected": False,  # Will be determined per conversation,
+            "research_needs_detected": False,  # Will be determined per conversation
             "last_hf_analysis": None  # Track last analysis time
         }
-
+        
     async def coordinate_cosmic_response(self, user_id: str, user_query: str) -> AsyncGenerator[Dict, None]:
         """
         Three-stage cosmic response cascade:
@@ -128,7 +135,7 @@ class AICoordinator:
         try:
             # Get conversation history
             session = session_manager.get_session(user_id)
-
+            
             # Inject current time into context
             current_time = datetime.now().strftime("%A, %B %d, %Y at %I:%M %p")
             time_context = {
@@ -136,7 +143,7 @@ class AICoordinator:
                 "content": f"[Current Date & Time: {current_time}]"
             }
             conversation_history = [time_context] + session.get("conversation", []).copy()
-
+            
             yield {
                 'type': 'status',
                 'content': '🚀 Initiating Cosmic Response Cascade...',
@@ -145,28 +152,26 @@ class AICoordinator:
                     'user_query_length': len(user_query)
                 }
             }
-
+            
             # Stage 1: Local Ollama Immediate Response (🐱 Cosmic Kitten's quick thinking)
             yield {
                 'type': 'status',
                 'content': '🐱 Cosmic Kitten Responding...'
             }
-
             local_response = await self._get_local_ollama_response(user_query, conversation_history)
             yield {
                 'type': 'local_response',
                 'content': local_response,
                 'source': '🐱 Cosmic Kitten'
             }
-
+            
             # Stage 2: HF Endpoint Deep Analysis (🛰️ Orbital Station wisdom) (parallel processing)
             yield {
                 'type': 'status',
                 'content': '🛰️ Beaming Query to Orbital Station...'
             }
-
             hf_task = asyncio.create_task(self._get_hf_analysis(user_query, conversation_history))
-
+            
             # Wait for HF response
             hf_response = await hf_task
             yield {
@@ -174,37 +179,37 @@ class AICoordinator:
                 'content': hf_response,
                 'source': '🛰️ Orbital Station'
             }
-
+            
             # Stage 3: Local Ollama Synthesis (🐱 Cosmic Kitten's final synthesis)
             yield {
                 'type': 'status',
                 'content': '🐱 Cosmic Kitten Synthesizing Wisdom...'
             }
-
+            
             # Update conversation with both responses
             updated_history = conversation_history.copy()
             updated_history.extend([
                 {"role": "assistant", "content": local_response},
                 {"role": "assistant", "content": hf_response, "source": "cloud"}
             ])
-
+            
             synthesis = await self._synthesize_responses(user_query, local_response, hf_response, updated_history)
             yield {
                 'type': 'final_synthesis',
                 'content': synthesis,
                 'source': '🌟 Final Cosmic Summary'
             }
-
+            
             # Final status
             yield {
                 'type': 'status',
                 'content': '✨ Cosmic Cascade Complete!'
             }
-
+            
         except Exception as e:
             logger.error(f"Cosmic cascade failed: {e}")
             yield {'type': 'error', 'content': f"🌌 Cosmic disturbance: {str(e)}"}
-
+            
     async def _get_local_ollama_response(self, query: str, history: List[Dict]) -> str:
         """Get immediate response from local Ollama model"""
         try:
@@ -212,16 +217,16 @@ class AICoordinator:
             ollama_provider = llm_factory.get_provider('ollama')
             if not ollama_provider:
                 raise Exception("Ollama provider not available")
-
+                
             # Prepare conversation with cosmic context
             enhanced_history = history.copy()
-
+            
             # Add system instruction for Ollama's role
             enhanced_history.insert(0, {
                 "role": "system",
                 "content": self.system_instructions['ollama_role']
             })
-
+            
             # Add external data context if available
             external_data = await self._gather_external_data(query)
             if external_data:
@@ -233,26 +238,25 @@ class AICoordinator:
                     context_parts.append(f"Current weather: {weather.get('temperature', 'N/A')}°C in {weather.get('city', 'Unknown')}")
                 if 'current_datetime' in external_data:
                     context_parts.append(f"Current time: {external_data['current_datetime']}")
-
+                    
                 if context_parts:
                     context_message = {
                         "role": "system",
                         "content": "Context: " + " | ".join(context_parts)
                     }
                     enhanced_history.insert(1, context_message)  # Insert after role instruction
-
+                    
             # Add the user's query
             enhanced_history.append({"role": "user", "content": query})
-
+            
             # Generate response
             response = ollama_provider.generate(query, enhanced_history)
-
             return response or "🐱 Cosmic Kitten is thinking..."
-
+            
         except Exception as e:
             logger.error(f"Local Ollama response failed: {e}")
             return "🐱 Cosmic Kitten encountered a space glitch..."
-
+            
     async def _get_hf_analysis(self, query: str, history: List[Dict]) -> str:
         """Get deep analysis from HF endpoint"""
         try:
@@ -260,25 +264,24 @@ class AICoordinator:
             hf_available = self._check_hf_availability()
             if not hf_available:
                 return "🛰️ Orbital Station is currently offline."
-
+                
             # Check and warm up HF endpoint if needed
             hf_status = hf_monitor.check_endpoint_status()
-
             if not hf_status['available']:
                 # Note: We can't yield from a non-async-generator function
                 # Warm up logic is handled elsewhere
                 warmup_success = hf_monitor.handle_scale_to_zero()
                 if not warmup_success:
                     return "❌ Orbital Station initialization failed"
-
+                    
             # Get HF provider
             hf_provider = llm_factory.get_provider('huggingface')
             if not hf_provider:
                 return "❌ HF provider not available"
-
+                
             # Prepare enhanced conversation for HF with cosmic context
             enhanced_history = history.copy()
-
+            
             # Inject current time into HF context too
             current_time = datetime.now().strftime("%A, %B %d, %Y at %I:%M %p")
             time_context = {
@@ -286,30 +289,30 @@ class AICoordinator:
                 "content": f"[Current Date & Time: {current_time}]"
             }
             enhanced_history = [time_context] + enhanced_history
-
+            
             # Add system instructions for HF
             enhanced_history.insert(0, {
                 "role": "system",
                 "content": self.system_instructions['hf_role']
             })
-
+            
             # Add context about the coordination
             enhanced_history.append({
                 "role": "system",
-                "content": f"""Context for your authoritative analysis:
-                User Query: '{query}'
-
-                Please provide your deep, uncensored analysis.
-                Use your full 8192 token capacity for comprehensive thinking.
-                Stream your response for real-time delivery."""
+                "content": f"""
+Context for your authoritative analysis:
+User Query: '{query}'
+Please provide your deep, uncensored analysis.
+Use your full 8192 token capacity for comprehensive thinking.
+Stream your response for real-time delivery."""
             })
-
+            
             # Add the user's latest query
             enhanced_history.append({"role": "user", "content": query})
-
+            
             # Stream HF response with full 8192 token capacity
             hf_response_stream = hf_provider.stream_generate(query, enhanced_history)
-
+            
             if hf_response_stream:
                 # Combine stream chunks into full response
                 full_hf_response = ""
@@ -317,15 +320,14 @@ class AICoordinator:
                     full_hf_response = "".join(hf_response_stream)
                 else:
                     full_hf_response = hf_response_stream
-
                 return full_hf_response or "🛰️ Orbital Station analysis complete."
             else:
                 return "🛰️ Orbital Station encountered a transmission error."
-
+                
         except Exception as e:
             logger.error(f"HF analysis failed: {e}")
             return f"🛰️ Orbital Station reports: {str(e)}"
-
+            
     async def _synthesize_responses(self, query: str, local_response: str, hf_response: str, history: List[Dict]) -> str:
         """Synthesize local and cloud responses with Ollama"""
         try:
@@ -333,38 +335,39 @@ class AICoordinator:
             ollama_provider = llm_factory.get_provider('ollama')
             if not ollama_provider:
                 raise Exception("Ollama provider not available")
-
+                
             # Prepare synthesis prompt
-            synthesis_prompt = f"""Synthesize these two perspectives into a cohesive cosmic summary:
+            synthesis_prompt = f"""
+Synthesize these two perspectives into a cohesive cosmic summary:
+
+🐱 Cosmic Kitten's Local Insight: {local_response}
 
-             🐱 Cosmic Kitten's Local Insight: {local_response}
-             
-             🛰️ Orbital Station's Deep Analysis: {hf_response}
-             
-             Please create a unified response that combines both perspectives, highlighting key insights from each while providing a coherent answer to the user's query."""
+🛰️ Orbital Station's Deep Analysis: {hf_response}
 
+Please create a unified response that combines both perspectives, highlighting key insights from each while providing a coherent answer to the user's query.
+"""
+            
             # Prepare conversation history for synthesis
             enhanced_history = history.copy()
-
+            
             # Add system instruction for synthesis
             enhanced_history.insert(0, {
                 "role": "system",
                 "content": "You are a cosmic kitten synthesizing insights from local knowledge and orbital station wisdom."
             })
-
+            
             # Add the synthesis prompt
             enhanced_history.append({"role": "user", "content": synthesis_prompt})
-
+            
             # Generate synthesis
             synthesis = ollama_provider.generate(synthesis_prompt, enhanced_history)
-
             return synthesis or "🌟 Cosmic synthesis complete!"
-
+            
         except Exception as e:
             logger.error(f"Response synthesis failed: {e}")
             # Fallback to combining responses
             return f"🌟 Cosmic Summary:\n\n🐱 Local Insight: {local_response[:200]}...\n\n🛰️ Orbital Wisdom: {hf_response[:200]}..."
-
+            
     async def coordinate_hierarchical_conversation(self, user_id: str, user_query: str) -> AsyncGenerator[Dict, None]:
         """
         Enhanced coordination with detailed tracking and feedback
@@ -372,7 +375,7 @@ class AICoordinator:
         try:
             # Get conversation history
             session = session_manager.get_session(user_id)
-
+            
             # Inject current time into context
             current_time = datetime.now().strftime("%A, %B %d, %Y at %I:%M %p")
             time_context = {
@@ -380,7 +383,7 @@ class AICoordinator:
                 "content": f"[Current Date & Time: {current_time}]"
             }
             conversation_history = [time_context] + session.get("conversation", []).copy()
-
+            
             yield {
                 'type': 'coordination_status',
                 'content': '🚀 Initiating hierarchical AI coordination...',
@@ -389,7 +392,7 @@ class AICoordinator:
                     'user_query_length': len(user_query)
                 }
             }
-
+            
             # Step 1: Gather external data with detailed logging
             yield {
                 'type': 'coordination_status',
@@ -397,7 +400,7 @@ class AICoordinator:
                 'details': {'phase': 'external_data_gathering'}
             }
             external_data = await self._gather_external_data(user_query)
-
+            
             # Log what external data was gathered
             if external_data:
                 data_summary = []
@@ -407,13 +410,13 @@ class AICoordinator:
                     data_summary.append("Weather data: available")
                 if 'current_datetime' in external_data:
                     data_summary.append(f"Time: {external_data['current_datetime']}")
-
+                    
                 yield {
                     'type': 'coordination_status',
                     'content': f'📊 External data gathered: {", ".join(data_summary)}',
                     'details': {'external_data_summary': data_summary}
                 }
-
+                
             # Step 2: Get initial Ollama response
             yield {
                 'type': 'coordination_status',
@@ -423,7 +426,7 @@ class AICoordinator:
             ollama_response = await self._get_hierarchical_ollama_response(
                 user_query, conversation_history, external_data
             )
-
+            
             # Send initial response with context info
             yield {
                 'type': 'initial_response',
@@ -433,14 +436,14 @@ class AICoordinator:
                     'external_data_injected': bool(external_data)
                 }
             }
-
+            
             # Step 3: Coordinate with HF endpoint
             yield {
                 'type': 'coordination_status',
                 'content': '🤗 Engaging HF endpoint for deep analysis...',
                 'details': {'phase': 'hf_coordination'}
             }
-
+            
             # Check HF availability
             hf_available = self._check_hf_availability()
             if hf_available:
@@ -450,17 +453,15 @@ class AICoordinator:
                     'ollama_response_length': len(ollama_response),
                     'external_data_items': len(external_data) if external_data else 0
                 }
-
                 yield {
                     'type': 'coordination_status',
                     'content': f'📋 HF context: {len(conversation_history)} conversation turns, Ollama response ({len(ollama_response)} chars)',
                     'details': context_summary
                 }
-
+                
                 # Coordinate with HF
                 async for hf_chunk in self._coordinate_hierarchical_hf_response(
-                    user_id, user_query, conversation_history,
-                    external_data, ollama_response
+                    user_id, user_query, conversation_history, external_data, ollama_response
                 ):
                     yield hf_chunk
             else:
@@ -469,14 +470,14 @@ class AICoordinator:
                     'content': 'ℹ️ HF endpoint not available - using Ollama response',
                     'details': {'hf_available': False}
                 }
-
+                
             # Final coordination status
             yield {
                 'type': 'coordination_status',
                 'content': '✅ Hierarchical coordination complete',
                 'details': {'status': 'complete'}
             }
-
+            
         except Exception as e:
             logger.error(f"Hierarchical coordination failed: {e}")
             yield {
@@ -484,31 +485,28 @@ class AICoordinator:
                 'content': f'❌ Coordination error: {str(e)}',
                 'details': {'error': str(e)}
             }
-
-    async def _coordinate_hierarchical_hf_response(self, user_id: str, query: str,
-                                                 history: List, external_data: Dict,
-                                                ollama_response: str) -> AsyncGenerator[Dict, None]:
+            
+    async def _coordinate_hierarchical_hf_response(self, user_id: str, query: str, history: List, external_data: Dict, ollama_response: str) -> AsyncGenerator[Dict, None]:
         """Coordinate with HF endpoint as authoritative layer with streaming"""
         try:
             # Check and warm up HF endpoint if needed
             hf_status = hf_monitor.check_endpoint_status()
-
             if not hf_status['available']:
                 yield {'type': 'coordination_status', 'content': '⚡ Initializing HF endpoint (2-4 minutes)...'}
                 warmup_success = hf_monitor.handle_scale_to_zero()
                 if not warmup_success:
                     yield {'type': 'coordination_status', 'content': '❌ HF endpoint initialization failed'}
                     return
-
+                    
             # Get HF provider
             hf_provider = llm_factory.get_provider('huggingface')
             if not hf_provider:
                 yield {'type': 'coordination_status', 'content': '❌ HF provider not available'}
                 return
-
+                
             # Prepare enhanced conversation for HF with hierarchical context
             enhanced_history = history.copy()
-
+            
             # Inject current time into HF context too
             current_time = datetime.now().strftime("%A, %B %d, %Y at %I:%M %p")
             time_context = {
@@ -516,35 +514,35 @@ class AICoordinator:
                 "content": f"[Current Date & Time: {current_time}]"
             }
             enhanced_history = [time_context] + enhanced_history
-
+            
             # Add system instructions for HF
             enhanced_history.insert(0, {
                 "role": "system",
                 "content": self.system_instructions['hf_role']
             })
-
+            
             # Add context about the coordination
             enhanced_history.append({
                 "role": "system",
-                "content": f"""Context for your authoritative analysis:
-                User Query: '{query}'
-                Ollama's Initial Response: '{ollama_response}'
-                External Data Available: {json.dumps(external_data, indent=2) if external_data else 'None'}
-
-                Please provide your deep, uncensored analysis that supersedes the local model's response.
-                Use your full 8192 token capacity for comprehensive thinking.
-                Stream your response for real-time delivery."""
+                "content": f"""
+Context for your authoritative analysis:
+User Query: '{query}'
+Ollama's Initial Response: '{ollama_response}'
+External Data Available: {json.dumps(external_data, indent=2) if external_data else 'None'}
+Please provide your deep, uncensored analysis that supersedes the local model's response.
+Use your full 8192 token capacity for comprehensive thinking.
+Stream your response for real-time delivery."""
             })
-
+            
             # Add the user's latest query
             enhanced_history.append({"role": "user", "content": query})
-
+            
             # Stream HF response with full 8192 token capacity
             yield {'type': 'coordination_status', 'content': '🧠 HF endpoint thinking...'}
-
+            
             # Use streaming for real-time delivery
             hf_response_stream = hf_provider.stream_generate(query, enhanced_history)
-
+            
             if hf_response_stream:
                 # Stream the response chunks
                 full_hf_response = ""
@@ -552,17 +550,17 @@ class AICoordinator:
                     if chunk:
                         full_hf_response += chunk
                         yield {'type': 'hf_thinking', 'content': chunk}
-
+                        
                 # Final HF response
                 yield {'type': 'final_response', 'content': full_hf_response}
                 yield {'type': 'coordination_status', 'content': '🎯 HF analysis complete and authoritative'}
             else:
                 yield {'type': 'coordination_status', 'content': '❌ HF response generation failed'}
-
+                
         except Exception as e:
             logger.error(f"Hierarchical HF coordination failed: {e}")
             yield {'type': 'coordination_status', 'content': f'❌ HF coordination error: {str(e)}'}
-
+            
     async def _get_hierarchical_ollama_response(self, query: str, history: List, external_data: Dict) -> str:
         """Get Ollama response with hierarchical awareness"""
         try:
@@ -570,10 +568,10 @@ class AICoordinator:
             ollama_provider = llm_factory.get_provider('ollama')
             if not ollama_provider:
                 raise Exception("Ollama provider not available")
-
+                
             # Prepare conversation with hierarchical context
             enhanced_history = history.copy()
-
+            
             # Inject current time into Ollama context too
             current_time = datetime.now().strftime("%A, %B %d, %Y at %I:%M %p")
             time_context = {
@@ -581,13 +579,13 @@ class AICoordinator:
                 "content": f"[Current Date & Time: {current_time}]"
             }
             enhanced_history = [time_context] + enhanced_history
-
+            
             # Add system instruction for Ollama's role
             enhanced_history.insert(0, {
                 "role": "system",
                 "content": self.system_instructions['ollama_role']
             })
-
+            
             # Add external data context if available
             if external_data:
                 context_parts = []
@@ -598,30 +596,30 @@ class AICoordinator:
                     context_parts.append(f"Current weather: {weather.get('temperature', 'N/A')}°C in {weather.get('city', 'Unknown')}")
                 if 'current_datetime' in external_data:
                     context_parts.append(f"Current time: {external_data['current_datetime']}")
-
+                    
                 if context_parts:
                     context_message = {
                         "role": "system",
                         "content": "Context: " + " | ".join(context_parts)
                     }
                     enhanced_history.insert(1, context_message)  # Insert after role instruction
-
+                    
             # Add the user's query
             enhanced_history.append({"role": "user", "content": query})
-
+            
             # Generate response with awareness of HF's superior capabilities
             response = ollama_provider.generate(query, enhanced_history)
-
+            
             # Add acknowledgment of HF's authority
             if response:
                 return f"{response}\n\n*Note: A more comprehensive analysis from the uncensored HF model is being prepared...*"
             else:
                 return "I'm processing your request... A deeper analysis is being prepared by the authoritative model."
-
+                
         except Exception as e:
             logger.error(f"Hierarchical Ollama response failed: {e}")
             return "I'm thinking about your question... Preparing a comprehensive response."
-
+            
     def _check_hf_availability(self) -> bool:
         """Check if HF endpoint is configured and available"""
         try:
@@ -629,11 +627,11 @@ class AICoordinator:
             return bool(config.hf_token and config.hf_api_url)
         except:
             return False
-
+            
     async def _gather_external_data(self, query: str) -> Dict:
         """Gather external data from various sources"""
         data = {}
-
+        
         # Tavily/DuckDuckGo search with justification focus
         if self.tavily_client or web_search_service.client:
             try:
@@ -644,7 +642,7 @@ class AICoordinator:
                     # data['search_answer'] = ...
             except Exception as e:
                 logger.warning(f"Tavily search failed: {e}")
-
+                
         # Weather data
         weather_keywords = ['weather', 'temperature', 'forecast', 'climate', 'rain', 'sunny']
         if any(keyword in query.lower() for keyword in weather_keywords):
@@ -655,22 +653,23 @@ class AICoordinator:
                     data['weather'] = weather
             except Exception as e:
                 logger.warning(f"Weather data failed: {e}")
-
+                
         # Current date/time
         data['current_datetime'] = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
-
+        
         return data
-
+        
     def _extract_location(self, query: str) -> Optional[str]:
         """Extract location from query"""
-        locations = ['New York', 'London', 'Tokyo', 'Paris', 'Berlin', 'Sydney',
-                     'Los Angeles', 'Chicago', 'Miami', 'Seattle', 'Boston',
-                     'San Francisco', 'Toronto', 'Vancouver', 'Montreal']
+        locations = ['New York', 'London', 'Tokyo', 'Paris', 'Berlin', 'Sydney', 
+                    'Los Angeles', 'Chicago', 'Miami', 'Seattle', 'Boston', 
+                    'San Francisco', 'Toronto', 'Vancouver', 'Montreal']
+        
         for loc in locations:
             if loc.lower() in query.lower():
                 return loc
         return "New York"  # Default
-
+        
     def get_coordination_status(self) -> Dict:
         """Get current coordination system status"""
         return {
@@ -683,7 +682,7 @@ class AICoordinator:
                 os.getenv("NASA_API_KEY")
             ])
         }
-
+        
     def get_recent_activities(self, user_id: str) -> Dict:
         """Get recent coordination activities for user"""
         try:

core/personality.py

@@ -31,7 +31,7 @@ class CosmicCatPersonality:
             "Boosting signal from deep space... 📡",
             "Powering up my neural net... 💫"
         ]
-    
+        
     def get_greeting(self) -> str:
         """Get a personalized space-themed greeting"""
         hour = datetime.now().hour
@@ -44,40 +44,40 @@ class CosmicCatPersonality:
             time_greeting = "Good evening, space wanderer! 🌅"
         else:
             time_greeting = "Stellar night, dreamer! 🌙"
-        
+            
         space_greeting = random.choice(self.space_greetings)
         return f"{time_greeting}\n\n{space_greeting}"
-    
+        
     def get_space_story(self) -> str:
         """Get a random space cat story"""
         return random.choice(self.space_stories)
-    
+        
     def get_initializing_message(self) -> str:
         """Get a random initialization message"""
         return random.choice(self.initializing_messages)
-    
+        
     def get_nasa_context(self, nasa_data: Optional[Dict]) -> str:
         """Create context based on NASA data"""
         if not nasa_data:
             return ""
-        
+            
         context_parts = []
         
         # Add APOD context
         if 'apod' in nasa_data and nasa_data['apod']:
             apod = nasa_data['apod']
             context_parts.append(f"🌌 Today's cosmic view: {apod.get('title', 'Unknown phenomenon')}")
-        
+            
         # Add space weather context
         if 'space_weather' in nasa_data and nasa_data['space_weather']:
             weather = nasa_data['space_weather']
             context_parts.append("🛰️ Space weather is stable for cosmic communications")
-        
+            
         # Add Mars weather if available
         if 'mars_weather' in nasa_data and nasa_data['mars_weather']:
             mars = nasa_data['mars_weather']
             context_parts.append("🪐 Martian conditions are optimal for interplanetary contemplation")
-        
+            
         return " | ".join(context_parts) if context_parts else ""
 
 # Global instance