Fix the intent issue

main.py

@@ -156,55 +156,37 @@ class HealthcareChatbot:
 
         # Intent classification prompt
         self.intent_classifier_template = PromptTemplate(
-            template="""
-            You are a strict intent classifier for a healthcare chatbot. Your task is to determine whether a user query requires an API call or is just a conversational message.
-            This decision MUST be based on an accurate comparison between the user's message and the functionality described in the available API endpoints.
-
-            === USER INPUT CONTEXT ===
-            User Message: {user_query}
-            Detected Language: {detected_language}
-            Conversation History: {conversation_history}
-
-            === API ENDPOINTS DOCUMENTATION ===
-            You must compare the user query against ALL of the following API endpoints and their descriptions to decide if there is a functional match:
-            {endpoints_documentation}
-
-            === DECISION INSTRUCTIONS ===
-            Step 1: Understand the intent and meaning of the user message.
-            Step 2: Carefully read all API endpoint descriptions.
-            Step 3: Check if the user query matches or requests a function explicitly supported by ANY of the endpoints.
-            - If there is a match, the intent is **API_ACTION**.
-            - If no endpoint matches the user query in purpose or meaning, the intent is **CONVERSATION**.
-
-            === RESPONSE FORMAT ===
-            Respond with ONLY the following JSON structure:
-            {{
-                "intent": "API_ACTION" or "CONVERSATION",
-                "confidence": [0.0 to 1.0],
-                "reasoning": "Explain if and how the user query matched any endpoint description, or why it didn’t",
-                "requires_backend": true or false
-            }}
-
-            === CLASSIFICATION RULES ===
-            Choose **API_ACTION** if the user request can be fulfilled using an existing endpoint. This includes:
-            - Scheduling, canceling, or managing appointments
-            - Requesting or viewing test results, prescriptions, or medical records
-            - Any interaction described in the API documentation
-
-            Choose **CONVERSATION** if the query:
-            - Is a greeting or casual message
-            - Asks about general health advice or medical definitions
-            - Mentions something unrelated to the provided API functionality
-            - Cannot be fulfilled by any described endpoint
-
-            === IMPORTANT NOTE ===
-            DO NOT guess. Only return "API_ACTION" if the user message CLEARLY maps to a specific endpoint. Otherwise, return "CONVERSATION".
-
-            === CLASSIFICATION START ===
-            Begin classification based on the above rules:
-            """,
-            input_variables=["user_query", "detected_language", "conversation_history", "endpoints_documentation"]
-        )
+          template="""
+          You are an intent classifier. Your job is simple: understand what the user wants and check if any API endpoint can do that.
+
+          User Message: {user_query}
+          Language: {detected_language}
+          API Endpoints: {endpoints_documentation}
+
+          Think step by step:
+
+          1. What does the user want from this message?
+          Read the user's message carefully. What is the user trying to say or accomplish? What would a human understand from this message?
+
+          2. Can any API endpoint fulfill what the user wants?
+          Look at each API endpoint. Does any endpoint do what the user is asking for? Be very precise - only say yes if there's a clear match.
+
+          Important rules:
+          - Focus ONLY on the current message, ignore conversation history for classification
+          - If the user is just talking, being social, or saying something casual, that's CONVERSATION
+          - Only choose API_ACTION if the user is clearly asking for something an API endpoint can do
+          - When you're not sure, choose CONVERSATION
+
+          Answer in this format:
+          {{
+              "intent": "API_ACTION" or "CONVERSATION",
+              "confidence": [0.0 to 1.0],
+              "reasoning": "What does the user want? Can any API do this?",
+              "requires_backend": true or false
+          }}
+          """,
+          input_variables=["user_query", "detected_language", "conversation_history", "endpoints_documentation"]
+      )