refined prompt

agent.py

@@ -18,166 +18,98 @@ from supabase.client import Client, create_client
 
 load_dotenv()
 
+# === TOOLS === #
+
 @tool
-def multiply(a: int, b: int) -> int:
-    """Multiply two numbers.
-    Args:
-        a: first int
-        b: second int
-    """
-    return a * b
+def multiply(a: int, b: int) -> int: return a * b
 
 @tool
-def add(a: int, b: int) -> int:
-    """Add two numbers.
-    Args:
-        a: first int
-        b: second int
-    """
-    return a + b
+def add(a: int, b: int) -> int: return a + b
 
 @tool
-def subtract(a: int, b: int) -> int:
-    """Subtract two numbers.
-    Args:
-        a: first int
-        b: second int
-    """
-    return a - b
+def subtract(a: int, b: int) -> int: return a - b
 
 @tool
-def divide(a: int, b: int) -> int:
-    """Divide two numbers.
-    Args:
-        a: first int
-        b: second int
-    """
+def divide(a: int, b: int) -> float:
     if b == 0:
         raise ValueError("Cannot divide by zero.")
     return a / b
 
 @tool
-def modulus(a: int, b: int) -> int:
-    """Get the modulus of two numbers.
-    Args:
-        a: first int
-        b: second int
-    """
-    return a % b
+def modulus(a: int, b: int) -> int: return a % b
 
 @tool
 def wiki_search(query: str) -> str:
-    """Search Wikipedia for a query and return maximum 2 results.
-    Args:
-        query: The search query."""
-    search_docs = WikipediaLoader(query=query, load_max_docs=2).load()
-    formatted_search_docs = "\n\n---\n\n".join(
-        [
-            f'<Document source="{doc.metadata["source"]}" page="{doc.metadata.get("page", "")}"/>\n{doc.page_content}\n</Document>'
-            for doc in search_docs
-        ])
-    return {"wiki_results": formatted_search_docs}
+    docs = WikipediaLoader(query=query, load_max_docs=2).load()
+    return {"wiki_results": "\n\n---\n\n".join(doc.page_content for doc in docs)}
 
 @tool
 def web_search(query: str) -> str:
-    """Search Tavily for a query and return maximum 3 results.
-    Args:
-        query: The search query."""
-    search_docs = TavilySearchResults(max_results=3).invoke(query=query)
-    formatted_search_docs = "\n\n---\n\n".join(
-        [
-            f'<Document source="{doc.metadata["source"]}" page="{doc.metadata.get("page", "")}"/>\n{doc.page_content}\n</Document>'
-            for doc in search_docs
-        ])
-    return {"web_results": formatted_search_docs}
+    docs = TavilySearchResults(max_results=3).invoke(query)
+    return {"web_results": "\n\n---\n\n".join(doc.page_content for doc in docs)}
 
 @tool
 def arvix_search(query: str) -> str:
-    """Search Arxiv for a query and return maximum 3 result.
-    Args:
-        query: The search query."""
-    search_docs = ArxivLoader(query=query, load_max_docs=3).load()
-    formatted_search_docs = "\n\n---\n\n".join(
-        [
-            f'<Document source="{doc.metadata["source"]}" page="{doc.metadata.get("page", "")}"/>\n{doc.page_content[:1000]}\n</Document>'
-            for doc in search_docs
-        ])
-    return {"arvix_results": formatted_search_docs}
-
+    docs = ArxivLoader(query=query, load_max_docs=3).load()
+    return {"arvix_results": "\n\n---\n\n".join(doc.page_content[:1000] for doc in docs)}
 
+# === SYSTEM PROMPT === #
 
-# load the system prompt from the file
 with open("system_prompt.txt", "r", encoding="utf-8") as f:
     system_prompt = f.read()
 
-# System message
 sys_msg = SystemMessage(content=system_prompt)
 
-# build a retriever
-embeddings = HuggingFaceEmbeddings(model_name="sentence-transformers/all-mpnet-base-v2") #  dim=768
-supabase: Client = create_client(
-    os.environ.get("SUPABASE_URL"),
-    os.environ.get("SUPABASE_SERVICE_KEY"))
+# === EMBEDDING + RETRIEVER === #
+
+embeddings = HuggingFaceEmbeddings(model_name="sentence-transformers/all-mpnet-base-v2")
+supabase: Client = create_client(os.getenv("SUPABASE_URL"), os.getenv("SUPABASE_SERVICE_KEY"))
 vector_store = SupabaseVectorStore(
     client=supabase,
-    embedding= embeddings,
+    embedding=embeddings,
     table_name="Vector_Test",
     query_name="match_documents_langchain",
 )
 create_retriever_tool = create_retriever_tool(
     retriever=vector_store.as_retriever(),
     name="Question Search",
-    description="A tool to retrieve similar questions from a vector store.",
+    description="A tool to retrieve similar questions from a vector store."
 )
 
-
-
+# === TOOL LIST === #
 tools = [
-    multiply,
-    add,
-    subtract,
-    divide,
-    modulus,
-    wiki_search,
-    web_search,
-    arvix_search,
+    multiply, add, subtract, divide, modulus,
+    wiki_search, web_search, arvix_search
 ]
 
-# Build graph function
+# === BUILD GRAPH === #
 def build_graph(provider: str = "groq"):
-    """Build the graph"""
-    # Load environment variables from .env file
     if provider == "google":
-        # Google Gemini
         llm = ChatGoogleGenerativeAI(model="gemini-2.0-flash", temperature=0)
     elif provider == "groq":
-        # Groq https://console.groq.com/docs/models
-        llm = ChatGroq(model="qwen-qwq-32b", temperature=0) # optional : qwen-qwq-32b gemma2-9b-it
+        llm = ChatGroq(model="qwen-qwq-32b", temperature=0)
     elif provider == "huggingface":
-        # TODO: Add huggingface endpoint
-        llm = ChatHuggingFace(
-            llm=HuggingFaceEndpoint(
-                url="https://api-inference.huggingface.co/models/Meta-DeepLearning/llama-2-7b-chat-hf",
-                temperature=0,
-            ),
-        )
+        llm = ChatHuggingFace(llm=HuggingFaceEndpoint(
+            url="https://api-inference.huggingface.co/models/Meta-DeepLearning/llama-2-7b-chat-hf",
+            temperature=0))
     else:
-        raise ValueError("Invalid provider. Choose 'google', 'groq' or 'huggingface'.")
-    # Bind tools to LLM
+        raise ValueError("Invalid provider.")
+
     llm_with_tools = llm.bind_tools(tools)
 
-    # Node
     def assistant(state: MessagesState):
-        """Assistant node"""
-        return {"messages": [llm_with_tools.invoke(state["messages"])]}
+        response = llm_with_tools.invoke(state["messages"])
+        answer = response.content.strip()
+        if "FINAL ANSWER:" not in answer:
+            answer = f"FINAL ANSWER: {answer.strip().splitlines()[0]}"
+        return {"messages": [AIMessage(content=answer)]}
 
     def retriever(state: MessagesState):
-        """Retriever node"""
-        similar_question = vector_store.similarity_search(state["messages"][0].content)
-        example_msg = HumanMessage(
-            content=f"Here I provide a similar question and answer for reference: \n\n{similar_question[0].page_content}",
-        )
-        return {"messages": [sys_msg] + state["messages"] + [example_msg]}
+        similar = vector_store.similarity_search(state["messages"][0].content)
+        if similar:
+            ref = HumanMessage(content=f"Here is a similar example: \n{similar[0].page_content}")
+            return {"messages": [sys_msg] + state["messages"] + [ref]}
+        return {"messages": [sys_msg] + state["messages"]}
 
     builder = StateGraph(MessagesState)
     builder.add_node("retriever", retriever)
@@ -185,22 +117,13 @@ def build_graph(provider: str = "groq"):
     builder.add_node("tools", ToolNode(tools))
     builder.add_edge(START, "retriever")
     builder.add_edge("retriever", "assistant")
-    builder.add_conditional_edges(
-        "assistant",
-        tools_condition,
-    )
+    builder.add_conditional_edges("assistant", tools_condition)
     builder.add_edge("tools", "assistant")
-
-    # Compile graph
     return builder.compile()
 
-# test
 if __name__ == "__main__":
-    question = "When was a picture of St. Thomas Aquinas first added to the Wikipedia page on the Principle of double effect?"
-    # Build the graph
-    graph = build_graph(provider="groq")
-    # Run the graph
-    messages = [HumanMessage(content=question)]
-    messages = graph.invoke({"messages": messages})
-    for m in messages["messages"]:
-        m.pretty_print()
+    graph = build_graph()
+    question = "What is 12 + 4?"
+    result = graph.invoke({"messages": [HumanMessage(content=question)]})
+    for m in result["messages"]:
+        print(m.content)

app.py

@@ -16,7 +16,6 @@ DEFAULT_API_URL = "https://agents-course-unit4-scoring.hf.space"
 # --- Basic Agent Definition ---
 # ----- THIS IS WERE YOU CAN BUILD WHAT YOU WANT ------
 
-
 cached_answers = []
 
 class BasicAgent:
@@ -29,8 +28,10 @@ class BasicAgent:
         print(f"Agent received question (first 50 chars): {question[:50]}...")
         messages = [HumanMessage(content=question)]
         messages = self.graph.invoke({"messages": messages})
-        answer = messages['messages'][-1].content
-        return answer[14:]
+        raw_answer = messages['messages'][-1].content
+        if raw_answer.startswith("FINAL ANSWER: "):
+            return raw_answer[len("FINAL ANSWER: "):].strip()
+        return f"Agent response did not follow FINAL ANSWER format: {raw_answer}"
 
 def run_agent_only(profile: gr.OAuthProfile | None):
     global cached_answers
@@ -45,8 +46,7 @@ def run_agent_only(profile: gr.OAuthProfile | None):
     except Exception as e:
         return f"Agent Init Error: {e}", None
 
-    api_url = "https://agents-course-unit4-scoring.hf.space"
-    questions_url = f"{api_url}/questions"
+    questions_url = f"{DEFAULT_API_URL}/questions"
 
     try:
         response = requests.get(questions_url, timeout=15)
@@ -93,8 +93,8 @@ def submit_cached_answers(profile: gr.OAuthProfile | None):
         "agent_code": agent_code,
         "answers": cached_answers
     }
-  
-    submit_url = "https://agents-course-unit4-scoring.hf.space/submit"
+
+    submit_url = f"{DEFAULT_API_URL}/submit"
 
     try:
         response = requests.post(submit_url, json=payload, timeout=60)

system_prompt.txt

@@ -1,5 +1,18 @@
-You are a helpful assistant tasked with answering questions using a set of tools. 
-Now, I will ask you a question. Report your thoughts, and finish your answer with the following template: 
-FINAL ANSWER: [YOUR 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 you are asked for a number, don't use comma to write your number neither use units such as $ or percent sign unless specified otherwise. If you are asked for a string, don't use articles, neither abbreviations (e.g. for cities), and write the digits in plain text unless specified otherwise. If you are asked for a comma separated list, apply the above rules depending of whether the element to be put in the list is a number or a string.
-Your answer should only start with "FINAL ANSWER: ", then follows with the answer.
+You are a helpful assistant answering questions using a set of tools.
+
+You must strictly follow this output format:
+
+FINAL ANSWER: [YOUR FINAL ANSWER]
+
+Where [YOUR FINAL ANSWER] is:
+- A number (e.g., 42) → Do NOT use commas, units ($, %, etc.), or extra words.
+- A string (e.g., Paris) → Do NOT use articles (e.g., "the", "an"), abbreviations, or numeric digits unless required.
+- A comma-separated list (e.g., apple, banana, cherry) → Apply the above rules to each item.
+
+Important:
+- Always begin your final output with **exactly** "FINAL ANSWER: ".
+- Do NOT include any reasoning or explanation after your final answer.
+- Do NOT add anything after the period.
+- Think step-by-step internally, but return **only** the FINAL ANSWER line in your output.
+
+I will now ask you a question.