added new system prompt

app.py

@@ -347,24 +347,32 @@ def run_and_submit_all(profile: gr.OAuthProfile | None):
 
 # --- Build Gradio Interface using Blocks ---
 with gr.Blocks() as demo:
-    gr.Markdown("# SmolAgent Evaluation Runner for GAIA")
+    gr.Markdown("# 🤖 SmolAgent GAIA Evaluation Runner")
     gr.Markdown(
         """
-        **Instructions:**
-
-        1. This agent uses smolagents framework with multiple tools:
-           - 🔍 **DuckDuckGoSearchTool**: Web search capabilities
-           - 🌐 **VisitWebpageTool**: Can visit and read web pages
-           - 🧮 **Math Calculator**: Handles mathematical calculations
-           - 📊 **Data Analysis**: Basic data analysis capabilities
-           - ✅ **Fact Checker**: Helps verify claims
-           - 🧠 **Advanced Reasoning**: Structured problem-solving
+        **Enhanced Agent for GAIA Dataset:**
+        
+        🛠️ **Tools Available:**
+        - 🔍 **DuckDuckGoSearchTool**: Real-time web search capabilities
+        - 🌐 **VisitWebpageTool**: Can visit and analyze web pages
+        - 🧮 **Math Calculator**: Safe mathematical calculations
+        - 📊 **Data Analysis**: Basic data analysis capabilities  
+        - ✅ **Fact Checker**: Helps verify claims with authoritative sources
+        - 🧠 **Advanced Reasoning**: Structured problem-solving approach
         
-        2. Log in to your Hugging Face account using the button below.
-        3. Click 'Run Evaluation & Submit All Answers' to start the evaluation.
+        🎯 **GAIA Format Compliance:**
+        - Numbers without commas or units (unless specified)
+        - Strings without articles or abbreviations
+        - Proper comma-separated lists
+        - Extracts only the final answer for submission
+
+        **Instructions:**
+        1. Log in to your Hugging Face account using the button below.
+        2. Click 'Run Evaluation & Submit All Answers' to start the evaluation.
+        3. The agent will process all questions using multiple tools and reasoning steps.
 
         ---
-        **Note:** This agent is designed for the GAIA dataset and can handle complex, multi-step reasoning tasks.
+        **Note:** This agent follows GAIA's strict answer formatting requirements and uses advanced reasoning with multiple tools.
         """
     )