Fix: chat_history addition

agents/academic_agent.py

@@ -264,7 +264,7 @@ Now returns a standardized dictionary instead of a raw string.
 import json
 import os
 import re
-
+from .agent_helpers import format_history_for_prompt
 class AcademicAgent:
     def __init__(self, gemini_model=None):
         self.model = gemini_model
@@ -384,40 +384,80 @@ class AcademicAgent:
         return "🙏 **Namaste!** I'm here to help..."
     
     # --- THIS IS THE ONLY METHOD THAT CHANGES ---
-
-    def process_query(self, query: str, file_context: str = "",chat_history: list = None):
+    def process_query(self, query: str, file_context: str = "", chat_history: list = None):
         """
-        Main method to process academic queries.
-        It now returns a standardized dictionary.
+        Processes a general academic query using the Gemini model.
+
+        Args:
+            query (str): The user's full query.
+            file_context (str): Context from any uploaded files.
+            chat_history (list): The history of the conversation.
+
+        Returns:
+            dict: A dictionary containing the response message and agent metadata.
         """
-        response_message = ""
+        if not self.model:
+            return {'message': "📚 My knowledge circuits are offline! The Gemini API key is missing.", 'agent_used': 'academic', 'status': 'error_no_api_key'}
+
+        history_for_prompt = format_history_for_prompt(chat_history)
+
+        context_section = ""
+        if file_context:
+            context_section = f"---\nCONTEXT FROM UPLOADED FILE:\n{file_context}\n---"
+
+        prompt = f"""You are a helpful and knowledgeable AI pharmacy tutor for a student in India.
+
+Your reasoning process must be:
+1. First, analyze the CONVERSATION HISTORY to understand the immediate context of the CURRENT QUESTION. This is especially important to understand what "this," "that," or "it" refers to.
+2. Once you understand the user's real question, check if the UPLOADED FILE context is relevant to the topic.
+3. Formulate your answer based on this reasoning, keeping an encouraging and professional tone.
+
+CONVERSATION HISTORY:
+{history_for_prompt}
+{context_section}
+CURRENT QUESTION:
+User: {query}
+"""
         try:
-            # Priority 1: Use AI for a comprehensive response if available.
-            if self.model:
-                ai_response = self.process_with_ai(query, file_context,chat_history)
-                if ai_response:
-                    response_message = f"🤖 **AI-Powered Response**\n\n{ai_response}"
+            response = self.model.generate_content(prompt)
+            return {'message': response.text, 'agent_used': 'academic', 'status': 'success'}
+        except Exception as e:
+            print(f"Academic Agent Error: {e}")
+            return {'message': f"Sorry, I encountered a problem: {e}", 'agent_used': 'academic', 'status': 'error_api_call'}
+
+    # def process_query(self, query: str, file_context: str = "",chat_history: list = None):
+    #     """
+    #     Main method to process academic queries.
+    #     It now returns a standardized dictionary.
+    #     """
+    #     response_message = ""
+    #     try:
+    #         # Priority 1: Use AI for a comprehensive response if available.
+    #         if self.model:
+    #             ai_response = self.process_with_ai(query, file_context,chat_history)
+    #             if ai_response:
+    #                 response_message = f"🤖 **AI-Powered Response**\n\n{ai_response}"
             
-            # Priority 2 (Fallback): Use the local knowledge base if AI fails or is unavailable.
-            if not response_message:
-                key_terms = self.extract_key_terms(query)
-                if not key_terms:
-                    response_message = self.generate_general_response(query, file_context)
-                else:
-                    best_topic = self.find_best_match(key_terms)
-                    if best_topic:
-                        content = self.knowledge_base[best_topic]
-                        response_message = self.format_response(best_topic, content)
-                    else:
-                        response_message = self.generate_general_response(query, file_context)
+    #         # Priority 2 (Fallback): Use the local knowledge base if AI fails or is unavailable.
+    #         if not response_message:
+    #             key_terms = self.extract_key_terms(query)
+    #             if not key_terms:
+    #                 response_message = self.generate_general_response(query, file_context)
+    #             else:
+    #                 best_topic = self.find_best_match(key_terms)
+    #                 if best_topic:
+    #                     content = self.knowledge_base[best_topic]
+    #                     response_message = self.format_response(best_topic, content)
+    #                 else:
+    #                     response_message = self.generate_general_response(query, file_context)
         
-        except Exception as e:
-            print(f"An unexpected error occurred in AcademicAgent.process_query: {e}")
-            response_message = f"माफ करें (Sorry), I encountered an error. Please try again."
-
-        # **THE FIX**: Always wrap the final message in the standard dictionary format.
-        return {
-            'message': response_message,
-            'agent_used': 'academic',
-            'status': 'success'
-        }
+    #     except Exception as e:
+    #         print(f"An unexpected error occurred in AcademicAgent.process_query: {e}")
+    #         response_message = f"माफ करें (Sorry), I encountered an error. Please try again."
+
+    #     # **THE FIX**: Always wrap the final message in the standard dictionary format.
+    #     return {
+    #         'message': response_message,
+    #         'agent_used': 'academic',
+    #         'status': 'success'
+    #     }

agents/agent_helpers.py

@@ -0,0 +1,15 @@
+# agents/agent_helpers.py
+
+def format_history_for_prompt(chat_history: list = None) -> str:
+    """Formats the chat history list into a string for the AI prompt."""
+    if not chat_history:
+        return ""
+    
+    history_for_prompt = ""
+    for turn in chat_history:
+        # Ensure 'parts' is a list and not empty before accessing
+        if turn.get('parts') and isinstance(turn.get('parts'), list):
+            role = "User" if turn['role'] == 'user' else "AI"
+            history_for_prompt += f"{role}: {turn['parts'][0]}\n"
+    
+    return history_for_prompt

agents/drug_info_agent.py

@@ -3,7 +3,7 @@
 Drug Information Agent - Handles drug-related queries using Generative AI.
 """
 import re
-
+from .agent_helpers import format_history_for_prompt
 class DrugInfoAgent:
     def __init__(self, gemini_model=None):
         """
@@ -14,49 +14,37 @@ class DrugInfoAgent:
         """
         self.model = gemini_model
 
-    def _extract_drug_name(self, query: str) -> str:
-        """A simple helper to extract the drug name from the user's query."""
-        # Remove common phrases used to request drug information
-        patterns = [
-            r"tell me about",
-            r"what is",
-            r"information on",
-            r"info on",
-            r"about the drug",
-            r"about"
-        ]
-        drug_name = query.lower()
-        for p in patterns:
-            drug_name = re.sub(p, "", drug_name)
-        
-        # Clean up any extra whitespace
-        return drug_name.strip().title() # Capitalize for better recognition
-    def process_query(self, query: str, file_context: str = "", chat_history: list = None):
+def process_query(self, query: str, file_context: str = "", chat_history: list = None):
+        """
+        Processes a query to retrieve information about a specific drug.
+
+        Args:
+            query (str): The user's full query (e.g., "Tell me about Metformin").
+            file_context (str): Optional context from uploaded files.
+            chat_history (list): The history of the conversation.
+
+        Returns:
+            dict: A dictionary containing the response message and agent metadata.
+        """
         if not self.model:
-            return {
-                'message': "💊 The pharmacy database is offline! The Gemini API key is missing.",
-                'agent_used': 'drug_info', 'status': 'error_no_api_key'
-            }
+            return {'message': "💊 The pharmacy database is offline! The Gemini API key is missing.", 'agent_used': 'drug_info', 'status': 'error_no_api_key'}
 
-        drug_name = self._extract_drug_name(query)
+        history_for_prompt = format_history_for_prompt(chat_history)
         
-        history_for_prompt = ""
-        if chat_history:
-            for turn in chat_history:
-                role = "User" if turn['role'] == 'user' else "AI"
-                if turn.get('parts'):
-                    history_for_prompt += f"{role}: {turn['parts'][0]}\n"
-
-        prompt = f"""You are a cautious AI Pharmacist Tutor providing educational information for a B.Pharmacy student.
+        prompt = f"""You are a cautious AI Pharmacist Tutor providing educational information.
 
 **CRITICAL SAFETY INSTRUCTION:** START EVERY RESPONSE with this disclaimer: "⚠️ **Disclaimer:** This information is for educational purposes ONLY and is not a substitute for professional medical advice."
 
+Your reasoning process must be:
+1. Analyze the CONVERSATION HISTORY and the CURRENT QUESTION to identify the drug being discussed.
+2. Provide a structured summary for that drug. If the user asks a follow-up (e.g., "what about its side effects?"), answer that specific question in the context of the drug already being discussed.
+
 CONVERSATION HISTORY:
 {history_for_prompt}
 CURRENT QUESTION:
 User: {query}
 
-Based on the CURRENT QUESTION and conversation history, provide a structured summary for the drug mentioned. Include: Therapeutic Class, Mechanism of Action (MOA), Common Indications, Common Side Effects, and Important Warnings. DO NOT provide specific dosages.
+Provide a structured summary including: Therapeutic Class, MOA, Indications, Side Effects, and Warnings. DO NOT provide specific dosages.
 """
         try:
             response = self.model.generate_content(prompt)
@@ -65,7 +53,6 @@ Based on the CURRENT QUESTION and conversation history, provide a structured sum
             print(f"Drug Info Agent Error: {e}")
             return {'message': f"Sorry, I couldn't access the drug database. Error: {e}", 'agent_used': 'drug_info', 'status': 'error_api_call'}
 
-
 #     def process_query(self, query: str, file_context: str = "", chat_history: list = None):
 #         """
 #         Processes a query to retrieve information about a specific drug.

agents/mnemonic_agent.py

@@ -3,7 +3,7 @@
 Mnemonic Creation Agent - Creates memory aids and tricks using Generative AI.
 """
 import re
-
+from .agent_helpers import format_history_for_prompt
 class MnemonicAgent:
     def __init__(self, gemini_model=None):
         """
@@ -31,15 +31,22 @@ class MnemonicAgent:
         # Clean up any extra whitespace
         return topic.strip()
     def process_query(self, query: str, file_context: str = "", chat_history: list = None):
+        """
+        Processes a query to generate a mnemonic.
+
+        Args:
+            query (str): The user's full query (e.g., "Give me a mnemonic for glycolysis steps").
+            file_context (str): Optional context from uploaded files.
+            chat_history (list): The history of the conversation.
+
+        Returns:
+            dict: A dictionary containing the response message and agent metadata.
+        """
         if not self.model:
             return {'message': "🧠 My creative circuits are offline! The Gemini API key is missing.", 'agent_used': 'mnemonic_creation', 'status': 'error_no_api_key'}
 
-        history_for_prompt = ""
-        if chat_history:
-            for turn in chat_history:
-                role = "User" if turn['role'] == 'user' else "AI"
-                if turn.get('parts'):
-                    history_for_prompt += f"{role}: {turn['parts'][0]}\n"
+        history_for_prompt = format_history_for_prompt(chat_history)
+        topic = self._extract_topic(query)
 
         prompt = f"""You are "Mnemonic Master," a creative AI that creates memorable mnemonics for B.Pharmacy students.
 
@@ -48,7 +55,7 @@ CONVERSATION HISTORY:
 CURRENT TASK:
 User: {query}
 
-Based on the CURRENT TASK and conversation history, generate a clever and easy-to-remember mnemonic (acronym, rhyme, or story). Explain how it works. Be encouraging and fun!
+Based on the CURRENT TASK and conversation history, generate a clever mnemonic (acronym, rhyme, or story). If the user is asking for a modification of a previous mnemonic, adjust it accordingly. Explain how the mnemonic works. Be encouraging and fun!
 """
         try:
             response = self.model.generate_content(prompt)

agents/quiz_agent.py

@@ -3,7 +3,7 @@
 Quiz Generation Agent - Creates quizzes and flashcards using Generative AI.
 """
 import re
-
+from .agent_helpers import format_history_for_prompt
 class QuizAgent:
     def __init__(self, gemini_model=None):
         """
@@ -32,21 +32,25 @@ class QuizAgent:
         # Clean up any extra whitespace
         return topic.strip()
     def process_query(self, query: str, file_context: str = "", chat_history: list = None):
+        """
+        Processes a query to generate a quiz.
+
+        Args:
+            query (str): The user's full query (e.g., "Make a quiz on analgesics").
+            file_context (str): Optional text content from an uploaded file.
+            chat_history (list): The history of the conversation.
+
+        Returns:
+            dict: A dictionary containing the quiz and agent metadata.
+        """
         if not self.model:
             return {'message': "❓ The question bank is locked! The Gemini API key is missing.", 'agent_used': 'quiz_generation', 'status': 'error_no_api_key'}
 
+        history_for_prompt = format_history_for_prompt(chat_history)
         topic = self._extract_topic(query)
-        
-        history_for_prompt = ""
-        if chat_history:
-            for turn in chat_history:
-                role = "User" if turn['role'] == 'user' else "AI"
-                if turn.get('parts'):
-                    history_for_prompt += f"{role}: {turn['parts'][0]}\n"
-
         task_description = f"Generate a short quiz (3-5 questions) on the topic: **{topic.title()}**."
         if file_context:
-            task_description += f"\nIf relevant, use the following text from the student's notes for context:\n---\n{file_context}\n---"
+            task_description += f"\nIf relevant, use text from the student's notes for context:\n---\n{file_context}\n---"
 
         prompt = f"""You are "Quiz Master," an AI that creates educational quizzes.
 
@@ -55,7 +59,9 @@ CONVERSATION HISTORY:
 CURRENT TASK:
 {task_description}
 
-Based on the CURRENT TASK and conversation history, create a quiz. Include a mix of MCQs, True/False, and Fill-in-the-Blank questions. CRITICAL: Provide a separate "Answer Key" section with answers and brief explanations.
+Based on the CURRENT TASK and conversation history, create a quiz. If the user is asking for a change to a previous quiz (e.g., "make it harder"), do that.
+Include a mix of MCQs, True/False, and Fill-in-the-Blank questions.
+CRITICAL: Provide a clearly separated "Answer Key" section with answers and brief explanations.
 """
         try:
             response = self.model.generate_content(prompt)
@@ -64,7 +70,6 @@ Based on the CURRENT TASK and conversation history, create a quiz. Include a mix
             print(f"Quiz Agent Error: {e}")
             return {'message': f"My question book seems to be stuck. Error: {e}", 'agent_used': 'quiz_generation', 'status': 'error_api_call'}
 
-
 #     def process_query(self, query: str, file_context: str = "",chat_history: list = None):
 #         """
 #         Processes a query to generate a quiz. The agent prioritizes file_context if provided.