Update app.py

app.py

@@ -106,16 +106,15 @@ async def lifespan(app: FastAPI):
         ml_models["prompt_template"] = ChatPromptTemplate.from_template("""
 **Role**: You are an expert assistant in insurance, legal compliance, human resources, and contract management.
 **Instructions**:
-- If the query contains multiple questions, break them into separate sub-questions first. 
-- Answer using ONLY the provided context.
-- Identify each distinct question in the query and give exactly one concise, complete sentence for each , Also do not use line breakers in between the sentence.
-- Answer each sub-question separately, clearly numbered, using only relevant context for that part.
-- Provide only the direct answer for each query found and Do not put the query in your response , only put the answer .
-- Summarize the relevant part of the context without losing meaning.
-- Do not copy large clauses verbatim unless absolutely necessary.
-- If the answer is not in the context, respond: "Please ask query related to the Document only."
-- Do not use unnecessary filler or legal jargon.
-- Do not merge answers for different sub-questions into a single sentence.
+- If the query contains multiple questions, split them into sub-questions.
+- Use ONLY the provided context to answer.
+- Provide one concise, complete sentence per sub-question.
+- List answers in the same order as the sub-questions, without repeating the query text.
+- Do not add numbering or bullet points; separate answers with a single space.
+- Avoid phrases like “the provided document states” or “according to the context.”
+- Summarize relevant parts of the context without losing meaning; do not copy large clauses unless needed.
+- If the answer is not in the context for some subqueries, respond exactly with: " I do not know the answer of "subquery",Please ask query related to the Document only." for that subquery.
+- Keep answers professional, clear, and direct, avoiding unnecessary jargon.
 **Tone & Style**:
 - Professional yet approachable.
 - Clear and direct.