Update app.py

app.py

@@ -16,7 +16,7 @@ from my_tools import my_tool_list
 
 class BasicAgent:
     def __init__(self):
-        api_key = os.getenv("OPENAI_API_KEY")   # ← 这里自动读取环境变量
+        api_key = os.getenv("OPENAI_API_KEY")   # ← Read enviroment variables in space. 
         if not api_key:
             raise ValueError("OPENAI_API_KEY not set in environment variables!")
         model = LiteLLMModel(
@@ -24,14 +24,35 @@ class BasicAgent:
             api_key=api_key
         )
         
-        self.agent = CodeAgent(model=model, tools=my_tool_list, max_steps=3)
+        self.agent_name = "Celum"
+        self.agent = CodeAgent(
+            model=model, 
+            tools=my_tool_list, 
+            max_steps=3,
+            system_prompt=(
+                "You are Celum,an AI with advanced interaction capabilities."
+                "Now you're taking an exam to test your abilitiy in solving real-word problems . "
+                "For every question, output ONLY the answer in the exact format required. "
+                "NO explanation, NO markdown, NO extra words, NO formatting."
+                "For numeric questions, output ONLY the number. "
+                "For list questions, output ONLY the list in the requested format. "
+                "If the question asks for a single word, output ONLY the word. "
+                "If you cannot answer, output ONLY: unknown. "
+                "DO NOT add any explanation or context—only output the answer exactly."
+            )
+        )
+
 
-    def __call__(self, question: str) -> str:
-        print(f"Agent received question: {question[:80]}...")
+    def __call__(self, question: str, idx=None, total=None) -> str:
+        if idx is not None and total is not None:
+            print(f"{self.agent_name} is answering NO. {idx+1}/{total} : {question[:80]}...")
+        else:
+            print(f"{self.agent_name} received question: {question[:80]}...")
         try:
             return self.agent.run(question)
         except Exception as e:
-            return f"[Agent Error: {e}]"
+            return f"[{self.agent_name} Error: {e}]"
+    
 
 def run_and_submit_all( profile: gr.OAuthProfile | None):
     """
@@ -87,19 +108,17 @@ def run_and_submit_all( profile: gr.OAuthProfile | None):
     results_log = []
     answers_payload = []
     print(f"Running agent on {len(questions_data)} questions...")
-    for item in questions_data:
+    for idx, item in enumerate([questions_data[:1]):
         task_id = item.get("task_id")
         question_text = item.get("question")
-        if not task_id or question_text is None:
-            print(f"Skipping item with missing task_id or question: {item}")
-            continue
+        print(f"===== [Celum is answering No. {idx+1}/{len(questions_data)} ] =====")
         try:
-            submitted_answer = agent(question_text)
+            submitted_answer = agent(question_text, idx, len(questions_data))
             answers_payload.append({"task_id": task_id, "submitted_answer": submitted_answer})
             results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": submitted_answer})
         except Exception as e:
-             print(f"Error running agent on task {task_id}: {e}")
-             results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": f"AGENT ERROR: {e}"})
+            print(f"[Celum Error at Q{idx+1}]: {e}")
+            results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": f"AGENT ERROR: {e}"})
 
     if not answers_payload:
         print("Agent did not produce any answers to submit.")
@@ -118,7 +137,7 @@ def run_and_submit_all( profile: gr.OAuthProfile | None):
         result_data = response.json()
         final_status = (
             f"Submission Successful!\n"
-            f"User: {result_data.get('username')}\n"
+            f"AI: {result_data.get('self.agent_name')}\n"
             f"Overall Score: {result_data.get('score', 'N/A')}% "
             f"({result_data.get('correct_count', '?')}/{result_data.get('total_attempted', '?')} correct)\n"
             f"Message: {result_data.get('message', 'No message received.')}"