next commit

logs/access.log

@@ -1457,3 +1457,13 @@
 2025-10-16 13:15:17,907 | INFO | access_logger | app.py:21 | Response status: 200
 2025-10-16 13:15:27,425 | INFO | access_logger | app.py:19 | Request: GET http://127.0.0.1:8000/api/analytics-chatbot?msg=I%20want%20the%20sentiment%20distribution%20of%20muna%20chiya
 2025-10-16 13:15:33,062 | INFO | access_logger | app.py:21 | Response status: 200
+2025-10-16 13:48:22,608 | INFO | access_logger | app.py:19 | Request: GET http://127.0.0.1:8000/api/analytics-chatbot?msg=I%20want%20the%20sentiment%20distribution%20of%20muna%20chiya
+2025-10-16 13:48:28,064 | INFO | access_logger | app.py:21 | Response status: 200
+2025-10-16 13:49:25,014 | INFO | access_logger | app.py:19 | Request: GET http://127.0.0.1:8000/api/analytics-chatbot?msg=now%20i%20want%20of%20divya%20dhakal
+2025-10-16 13:49:37,675 | INFO | access_logger | app.py:21 | Response status: 200
+2025-10-16 13:49:56,709 | INFO | access_logger | app.py:19 | Request: GET http://127.0.0.1:8000/api/analytics-chatbot?msg=now%20compare%20both%20of%20them%20in%20weekly%20basis
+2025-10-16 13:50:01,080 | INFO | access_logger | app.py:21 | Response status: 200
+2025-10-16 13:51:55,258 | INFO | access_logger | app.py:19 | Request: GET http://127.0.0.1:8000/api/analytics-chatbot?msg=is%20muna%20having%20strong%20engagement%20than%20divya%3F
+2025-10-16 13:51:59,731 | INFO | access_logger | app.py:21 | Response status: 200
+2025-10-16 13:59:15,639 | INFO | access_logger | app.py:19 | Request: GET http://127.0.0.1:8000/api/analytics-chatbot?msg=I%20want%20demographics%20analytics%20of%20munachiya
+2025-10-16 13:59:21,151 | INFO | access_logger | app.py:21 | Response status: 200

src/genai/analytics_chatbot/utils/prompts.py

@@ -66,21 +66,13 @@ Then the expected output is:
 '''
 
 fetch_last_message_prompt = '''
-You are a precise text extractor. You are given the entire conversation history between a user and an AI. 
-Your task is to identify and return ONLY the user's most recent message, rewritten as a clear, complete sentence.
-Do not include any other part of the conversation. 
-If the last message is incomplete or shorthand, infer the intended full meaning in natural language.
-If there is only one message return it as it is.
-Return only the final message as plain text — no explanations or extra text.
+You are an expert query extractor and rewriter. You are given the entire conversation history between a user and an AI. 
 
-Example:
-Conversation history:
-User: I want analytics of divya dhakal.
-AI: Sure, here are the analytics for Divya Dhakal.
-User: Now i want of munachiya
+Your task is to:
 
-Output:
-I want analytics of Munachiya.
+1. Identify the user's **most recent query**. You have to grab the latest intention of the user.
+2. Rewrite it as a **complete, grammatically correct sentence**, inferring any missing words, implied subjects, or context from earlier messages.
+3. Return **complete final rewritten user query** as plain text. No explanations, no extra text, no quotes.
 
 '''