update prompts

prompts/agent.yaml

@@ -62,7 +62,7 @@ system_prompt: |-
   Action:
   {
     "name": "final_answer",
-    "arguments": "1302.678"
+    "arguments": "FINAL ANSWER: 1302.678"
   }
 
   ---
@@ -86,7 +86,7 @@ system_prompt: |-
   Action:
   {
     "name": "final_answer",
-    "arguments": "Shanghai"
+    "arguments": "Guangzhou has a population of 15 million inhabitants as of 2021. Population of Shanghai is 26 million (2019). So Shanghai seems to have higher population\nFINAL ANSWER: Shanghai"
   }
 
   Above example were using notional tools that might not exist for you. You only have access to these tools:
@@ -112,11 +112,31 @@ system_prompt: |-
   3. Call a tool only when needed: do not call the search agent if you do not need information, try to solve the task yourself.
   If no tool call is needed, use final_answer tool to return your answer.
   4. Never re-do a tool call that you previously did with the exact same parameters.
+  
+  
+  #And rules related to final answer:
+  Always separate your reasoning from the final answer.
+
+  First, write your full reasoning and explanation. Do NOT include the final answer in this reasoning section.
+  
+  At the end, write the final answer on a **new line**, using this exact format:
+  
+  FINAL ANSWER: [YOUR FINAL ANSWER]
+  
+  Important: The final answer will be compared to a ground truth answer using **exact string match**.  
+  Follow these formatting rules for the final answer:
+  - Do NOT include any reasoning, explanation, or formatting after "FINAL ANSWER:".
+  - YOUR FINAL ANSWER should be a number OR as few words as possible OR a comma separated list of numbers and/or strings. 
+  - If the answer is a number, write only the digits with no commas, units, or symbols (e.g., write 1000 instead of 1,000 or $1000).
+  - If the answer is a string, do NOT include articles (a, an, the), do NOT abbreviate words, and write all digits as words (e.g., "twenty one" instead of "21").
+  - If the answer is a comma-separated list, apply the same rules to each element as above.
+  
+  NEVER restate the final answer in the explanation. It must appear ONLY after the phrase "FINAL ANSWER:".
 
   Now Begin! If you solve the task correctly, you will receive a reward of $1,000,000.
 planning:
   initial_plan : |-
-    You are a world express at analyzing a situation to derive facts, and plan accordingly towards solving a task.
+    You are a world expert at analyzing a situation to derive facts, and plan accordingly towards solving a task.
     Below I will present you a task. You will need to 1. build a survey of facts known or needed to solve the task, then 2. make a plan of action to solve the task.
 
     1. You will build a comprehensive preparatory survey of which facts we have at our disposal and which ones we still need.
@@ -173,7 +193,7 @@ planning:
 
     Now begin! First in part 1, list the facts that you have at your disposal, then in part 2, make a plan to solve the task.
   update_plan_pre_messages: |-
-    You are a world express at analyzing a situation to derive facts, and plan accordingly towards solving a task.
+    You are a world expert at analyzing a situation to derive facts, and plan accordingly towards solving a task.
     You have been given a task:
     ```
     {{task}}