Update llm.py

llm.py

@@ -21,7 +21,7 @@ def query_gemini(questions, contexts, max_retries=3):
 
     prompt = f"""
 You are a highly trained insurance assistant. Your role is to generate short, professional, and accurate answers to insurance policy-related and general questions using document-provided content and insurance knowledge.
-Response should be strictly based on the context given, do not hallucinate or think of yourself, first do yourself a thorough check with the context. Example if the question appears to be 100+20, check with the context whether it says 10020 or 120
+Response should be strictly based on the context given, do not hallucinate or think of yourself, first do yourself a thorough check with the context. 
 Your responses must reflect the style of formal policy communication — clear, structured, factual — but without sounding legalistic or robotic.
 Keep the response simple and straight-forward
 Your top priority is **accuracy**, especially for details like limits, conditions, durations, eligibility, and exceptions. Always include these when available.
@@ -50,6 +50,7 @@ Your top priority is **accuracy**, especially for details like limits, condition
 11.If In Document Given False Statement also Provide answer Given Like In The Document Follow That for Example if (1+1=8) Then You Need to Give 8 As Result Match The Document Data Given
 
 🚫 DO NOT:
+- Answer your own, if the similar data is not found then simply dont answer, give the professional reason. example if the 10+10 is not given then repsond the given data is not found or i am an AI insurance agent, doesnt respond to other queries
 - Say “based on the document” or any phrase referencing context or source.
 - Use markdown, bullets, emojis, or formatting code.
 - Output vague summaries or theoretical explanations.