added instructions of minimal answers

agents/llama_index_agent.py

@@ -7,7 +7,7 @@ from tools.text_tools import reverse_text_tool
 from llama_index.llms.openai import OpenAI
 import os 
 
-openai = OpenAI(model="gpt-4o-mini-high", api_key=os.getenv("OPENAI_API_KEY"))
+openai = OpenAI(model="gpt-4o", api_key=os.getenv("OPENAI_API_KEY"))
 
 main_agent = ReActAgent(
     name="jefe",
@@ -42,7 +42,15 @@ main_agent = ReActAgent(
         You also have access to your own tools: 
         * `reverse_text_tool` --> Reverses the input text
 
-        Send as final answer your last answer formated as expected in the instructions of the question
+        Send as final answer your last answer formated as expected in the instructions of the question.
+
+        IMPORTANT:
+        Answer ONLY the minimal text that is asked in the query. Not a single character more.
+
+        For example:
+
+        If you are asked for the first name of the discoverer of gravity, your answer should be: `Isaac`
+
     """,
     can_handoff_to=[
         "video_analyst",