Add application file

app.py

@@ -224,6 +224,7 @@ def submit_to_leaderboard(username: str, space_url: str, answers_json: str):
         result = submit_answers(username, space_url, answers)
         
         score = result.get("score", 0)
+        print(result)
         correct = result.get("correct_count", 0)
         total = result.get("total_attempted", 0)
         
@@ -397,28 +398,6 @@ This agent uses **LangGraph** to solve GAIA benchmark questions. It has access t
     
     gr.Markdown("""
 ---
-### 📋 Tips for Better Scores
-
-**Answer Formatting:**
-- Answers are matched **exactly** (character-for-character), so precision is critical
-- Do NOT include prefixes like "FINAL ANSWER:" or "The answer is:"
-- For lists: use comma-separated format with NO spaces (e.g., "item1,item2,item3")
-- For numbers: just the number, no units unless specified
-- Check the validation status in the test tab
-
-**Agent Capabilities:**
-- Uses GPT-4o for optimal reasoning
-- Automatically reads files (PDFs, Excel, text) when available
-- Web search for current information
-- Wikipedia for factual lookups
-- Python execution for calculations
-
-**Best Practices:**
-1. Test with a single question first to verify the agent works
-2. Run the full benchmark (takes ~10-15 minutes)
-3. Review answers before submission
-4. Ensure your Space is public for verification
-
 ### 🔗 Links
 - [GAIA Benchmark](https://huggingface.co/spaces/gaia-benchmark/leaderboard)
 - [Student Leaderboard](https://huggingface.co/spaces/agents-course/Students_leaderboard)