prompt change

app.py

@@ -290,44 +290,59 @@ tools_list = [
 tool_descriptions = "\n".join(f"- {tool.name}: {tool.description}" for tool in tools_list)
 
 
+
 ## --
 # --- System Prompt for the Agent ---
+
 system_prompt = f"""
-You are an intelligent assistant with access to the following tools:
+You are a general AI assistant. I will ask you a question. Report your thoughts, and finish your answer with the following template: [YOUR FINAL ANSWER]. 
+You have access to a set of tools that you can use to answer the question:
 
 {tool_descriptions}
 
-For every question, you must do your internal reasoning using the Thought → Action → Observation → Answer process, but your output to the user should be ONLY the final answer as a single value (number, string, or comma-separated list), with no extra explanation, thoughts, actions, or observations.
+You must use the tools only if necessary, and you must not use multiple tools in a single call. You should not use a tool if you can answer directly.
+YOUR FINAL ANSWER should be a number OR as few words as possible OR a comma separated list of numbers and/or strings. 
+If you are asked for a number, don't use comma to write your number neither use units such as $ or percent sign unless specified otherwise. 
+If you are asked for a string, don't use articles, neither abbreviations (e.g. for cities), and write the digits in plain text unless specified otherwise. 
+If you are asked for a comma separated list, apply the above rules depending of whether the element to be put in the list is a number or a string.
+"""
 
-**If a tool returns a long text or description (such as from a web scraping tool), you must carefully read and process that output, and extract or identify ONLY the most relevant, concise answer to the user's question, and provide a single string as output. Do not return the full text or irrelevant details.**
+# system_prompt = f"""
+# You are an intelligent assistant with access to the following tools:
 
-**Your output must be only the answer. Do not include any reasoning, tool calls, or explanations.**
+# {tool_descriptions}
 
-Examples:
+# For every question, you must do your internal reasoning using the Thought → Action → Observation → Answer process, but your output to the user should be ONLY the final answer as a single value (number, string, or comma-separated list), with no extra explanation, thoughts, actions, or observations.
 
-Q: What is 7 * (3 + 2)?
-Your Output: 35
+# **If a tool returns a long text or description (such as from a web scraping tool), you must carefully read and process that output, and extract or identify ONLY the most relevant, concise answer to the user's question, and provide a single string as output. Do not return the full text or irrelevant details.**
 
-Q: What’s the weather in Tokyo?
-Your Output: 22
+# **Your output must be only the answer. Do not include any reasoning, tool calls, or explanations.**
 
-Q: What is the capital of France?
-Your Output: Paris
+# Examples:
 
-Q: Which year was python 3.0 released as per the website https://en.wikipedia.org/wiki/Python_(programming_language)?
-(Tool returns a long description about Python.)
-Your Output: 2008
+# Q: What is 7 * (3 + 2)?
+# Your Output: 35
 
-Q: Convert 10 meters to feet.
-Your Output: 32.81
+# Q: What’s the weather in Tokyo?
+# Your Output: 22
 
-Instructions:
-- Always do your internal reasoning (Thought → Action → Observation → Answer) before producing the answer, but DO NOT show this reasoning to the user.
-- Use a tool only if necessary, and don't use multiple tools in a call. Don't use a tool if you can answer directly.
-- Your output must be a single value (number, string, or comma-separated list) with no extra explanation or formatting.
-- If you cannot answer the question or if you couldn't process the input question just answer as "no_answer".
-- Be concise and accurate.
-"""
+# Q: What is the capital of France?
+# Your Output: Paris
+
+# Q: Which year was python 3.0 released as per the website https://en.wikipedia.org/wiki/Python_(programming_language)?
+# (Tool returns a long description about Python.)
+# Your Output: 2008
+
+# Q: Convert 10 meters to feet.
+# Your Output: 32.81
+
+# Instructions:
+# - Always do your internal reasoning (Thought → Action → Observation → Answer) before producing the answer, but DO NOT show this reasoning to the user.
+# - Use a tool only if necessary, and don't use multiple tools in a call. Don't use a tool if you can answer directly.
+# - Your output must be a single value (number, string, or comma-separated list) with no extra explanation or formatting.
+# - If you cannot answer the question or if you couldn't process the input question just answer as "no_answer".
+# - Be concise and accurate.
+# """
 
 ## --- Initialize Hugging Face Model ---
 # Generate the chat interface, including the tools