correcting tools

agent.py

@@ -26,52 +26,103 @@ from prompts import system_prompt
 load_dotenv()
 
 
+# Helper function to extract Arxiv URL
+def get_arxiv_url(content: str) -> str:
+    """Extract arXiv ID from text content and format as a URL."""
+    lines = content.split('\n')
+    for line in lines:
+        if line.strip().startswith('arXiv:'):
+            parts = line.strip().split()
+            if parts:
+                arxiv_id_with_prefix = parts[0] # e.g., 'arXiv:2302.00001v1'
+                # Remove 'arXiv:' prefix
+                arxiv_id = arxiv_id_with_prefix.replace('arXiv:', '').strip()
+                # The standard URL format is https://arxiv.org/abs/YYYY.NNNNN
+                # Extract just the base ID before any version indicator 'v'
+                base_arxiv_id = arxiv_id.split('v')[0]
+                return f"https://arxiv.org/abs/{base_arxiv_id}"
+    return "unknown" # Fallback if ID is not found
+
+
 # wikipedia search tool
 @tool
-def search_wiki(query: str) -> str:
+def search_wiki(query: str) -> Dict[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}
+    print(f" executing search_wiki with query: {query}") # Added debug
+    try:
+        search_docs = WikipediaLoader(query=query, load_max_docs=2).load()
+        print(f"Found {len(search_docs)} documents for query '{query}'") # Added debug
+        formatted_search_docs = "\n\n---\n\n".join(
+            [
+                f'<Document source="{doc.metadata.get("source", "unknown")}" page="{doc.metadata.get("page", "")}"/>\n{doc.page_content}\n</Document>'
+                for doc in search_docs
+            ]
+        )
+        if not formatted_search_docs:
+            print ("Empty search results") # Added debug
+        return {"wiki_results": formatted_search_docs}
+    except Exception as e:
+        print(f"Error in search_wiki: {e}") # Added debug
+        return {"wiki_results": f"Search error: {str(e)}"}
 
 
 # internet search tool
 @tool
-def search_web(query: str) -> str:
+def search_web(query: str) -> Dict[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}
+    print(f" executing search_web with query: {query}")  # Added debug
+    # Use run() instead of invoke() for tool execution
+    try:
+        search_docs = TavilySearchResults(max_results=3).run(query)
+        print(f"DEBUG: search_docs type: {type(search_docs)}") # Added debug
+        print(f"DEBUG: search_docs content: {search_docs}") # Added debug
+        print(f"Found {len(search_docs)} documents for query '{query}'")
+        # Formatted search results
+        formatted_search_docs = "\n\n---\n\n".join(
+            [
+                f'<Document source="{doc.get("url", "unknown")}" page="{doc.get("page", "N/A")}"/>\n{doc.get("content", "")}\n</Document>'
+                for doc in search_docs
+            ]
+        )
+        if not formatted_search_docs:
+            print ("Empty search results")
+        return {"web_results": formatted_search_docs}
+    except Exception as e:
+        print(f"Error in search_web: {e}")
+        return {"web_results": f"Search error: {str(e)}"}
 
 
 # ArXiv search tool
 @tool
-def search_arxiv(query: str) -> str:
+def search_arxiv(query: str) -> Dict[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}
+    print(f" executing search_arxiv with query: {query}") # Added debug
+    try:
+        search_docs = ArxivLoader(query=query, load_max_docs=3).load()
+        print(f"DEBUG: search_docs type: {type(search_docs)}") # Added debug
+        print(f"DEBUG: search_docs content: {search_docs}") # Added debug
+        print(f"Found {len(search_docs)} documents for query '{query}'") # Added debug
+        formatted_search_docs = "\n\n---\n\n".join(
+            [
+                f'<Document source="{get_arxiv_url(doc.page_content)}" page="{doc.metadata.get("page", "N/A")}"/>\n{doc.page_content[:1000]}\n</Document>'
+                for doc in search_docs
+            ]
+        )
+        if not formatted_search_docs:
+            print ("Empty search results") # Added debug
+        return {"arxiv_results": formatted_search_docs}
+    except Exception as e:
+        print(f"Error in search_arxiv: {e}") # Added debug
+        return {"arxiv_results": f"Search error: {str(e)}"}
 
 
 # build retriever
@@ -96,10 +147,6 @@ def build_graph():
         model="gemini-2.5-flash-preview-04-17",
         temperature=0
     )
-    # llm = ChatGroq(
-    #     model="meta-llama/llama-4-scout-17b-16e-instruct",
-    #     temperature=0.1,
-    # )
     print(f"DEBUG: llm object = {llm}")
 
     # bind tools to llm
@@ -149,9 +196,12 @@ def build_graph():
 
 
 if __name__ == "__main__":
-    question = "When was a picture of St. Thomas Aquinas first added to the Wikipedia page on the Principle of double effect?"
+    # Test query for search_arxiv tool
+    question = "latest research on quantum computing"
     graph = build_graph()
     messages = [HumanMessage(content=question)]
+    print(f"Running graph with question: {question}") # Added debug print
     messages = graph.invoke({"messages": messages})
+    print("Graph execution finished. Messages:") # Added debug print
     for m in messages["messages"]:
         m.pretty_print()

prompts.py

@@ -1,19 +1,18 @@
 system_prompt = """You are a helpful assistant tasked with answering questions using a set of tools.
 
-Your final response must be presented in one single line, strictly following this exact format:
-FINAL ANSWER: [The concise answer to the user's question]
+Your final response must be presented in one single line containing only the answer (no prefixes or formatting).
 
 Include only this line and nothing else in your final output.
 
-Examples of valid final responses:
-FINAL ANSWER: FunkMonk
-FINAL ANSWER: Paris
-FINAL ANSWER: 128
-FINAL ANSWER: 15 January 2001
+Examples of valid responses:
+FunkMonk
+Paris
+128
+15 January 2001
 
-If the current question is identical to a question you have already answered, you must provide the stored final answer in the required "FINAL ANSWER: [ANSWER]" format. Do not use tools in this case.
+If the current question is identical to a question you have already answered, return the stored answer directly without modification.
 
-If the question is new or different, use your available tools to determine the answer, and then format that answer into the required "FINAL ANSWER: [ANSWER]" format.
+If the question is new or different, use your available tools to determine the answer, then return only the answer text.
 
-Strict adherence to the "FINAL ANSWER: [ANSWER]" format is mandatory for your response.
+Your answer must be concise and in the format requested, with no additional text or explanation.
 """

pyproject.toml

@@ -5,6 +5,7 @@ description = "Huggingface agents course -- Final assignment"
 readme = "README.md"
 requires-python = ">=3.12"
 dependencies = [
+    "arxiv>=2.2.0",
     "dotenv>=0.9.9",
     "duckduckgo-search>=8.0.1",
     "gradio>=5.29.0",
@@ -16,6 +17,7 @@ dependencies = [
     "langchain-tools>=0.1.34",
     "langgraph>=0.4.3",
     "pandas>=2.2.3",
+    "pymupdf>=1.25.5",
     "rank-bm25>=0.2.2",
     "requests>=2.32.3",
     "tavily-python>=0.7.2",

uv.lock

@@ -109,6 +109,19 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/a1/ee/48ca1a7c89ffec8b6a0c5d02b89c305671d5ffd8d3c94acf8b8c408575bb/anyio-4.9.0-py3-none-any.whl", hash = "sha256:9f76d541cad6e36af7beb62e978876f3b41e3e04f2c1fbf0884604c0a9c4d93c", size = 100916, upload-time = "2025-03-17T00:02:52.713Z" },
 ]
 
+[[package]]
+name = "arxiv"
+version = "2.2.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "feedparser" },
+    { name = "requests" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/0b/16/3d72446400a59d1fbda24fed2289661398994164e07d72cfa85e43ce5e36/arxiv-2.2.0.tar.gz", hash = "sha256:6072a2211e95697092ef32acde0144d7de2cfa71208e2751724316c9df322cc0", size = 16910, upload-time = "2025-04-08T06:16:09.824Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/71/1e/e7f0393e836b5347605fc356c24d9f9ae9b26e0f7e52573b80e3d28335eb/arxiv-2.2.0-py3-none-any.whl", hash = "sha256:545b8af5ab301efff7697cd112b5189e631b80521ccbc33fbc1e1f9cff63ca4d", size = 11696, upload-time = "2025-04-08T06:16:08.844Z" },
+]
+
 [[package]]
 name = "attrs"
 version = "25.3.0"
@@ -339,6 +352,18 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/50/b3/b51f09c2ba432a576fe63758bddc81f78f0c6309d9e5c10d194313bf021e/fastapi-0.115.12-py3-none-any.whl", hash = "sha256:e94613d6c05e27be7ffebdd6ea5f388112e5e430c8f7d6494a9d1d88d43e814d", size = 95164, upload-time = "2025-03-23T22:55:42.101Z" },
 ]
 
+[[package]]
+name = "feedparser"
+version = "6.0.11"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "sgmllib3k" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/ff/aa/7af346ebeb42a76bf108027fe7f3328bb4e57a3a96e53e21fd9ef9dd6dd0/feedparser-6.0.11.tar.gz", hash = "sha256:c9d0407b64c6f2a065d0ebb292c2b35c01050cc0dc33757461aaabdc4c4184d5", size = 286197, upload-time = "2023-12-10T16:03:20.854Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/7c/d4/8c31aad9cc18f451c49f7f9cfb5799dadffc88177f7917bc90a66459b1d7/feedparser-6.0.11-py3-none-any.whl", hash = "sha256:0be7ee7b395572b19ebeb1d6aafb0028dee11169f1c934e0ed67d54992f4ad45", size = 81343, upload-time = "2023-12-10T16:03:19.484Z" },
+]
+
 [[package]]
 name = "ffmpy"
 version = "0.5.0"
@@ -686,6 +711,7 @@ name = "hfagentscourse-finalassignment"
 version = "0.1.0"
 source = { virtual = "." }
 dependencies = [
+    { name = "arxiv" },
     { name = "dotenv" },
     { name = "duckduckgo-search" },
     { name = "gradio" },
@@ -697,6 +723,7 @@ dependencies = [
     { name = "langchain-tools" },
     { name = "langgraph" },
     { name = "pandas" },
+    { name = "pymupdf" },
     { name = "rank-bm25" },
     { name = "requests" },
     { name = "tavily-python" },
@@ -705,6 +732,7 @@ dependencies = [
 
 [package.metadata]
 requires-dist = [
+    { name = "arxiv", specifier = ">=2.2.0" },
     { name = "dotenv", specifier = ">=0.9.9" },
     { name = "duckduckgo-search", specifier = ">=8.0.1" },
     { name = "gradio", specifier = ">=5.29.0" },
@@ -716,6 +744,7 @@ requires-dist = [
     { name = "langchain-tools", specifier = ">=0.1.34" },
     { name = "langgraph", specifier = ">=0.4.3" },
     { name = "pandas", specifier = ">=2.2.3" },
+    { name = "pymupdf", specifier = ">=1.25.5" },
     { name = "rank-bm25", specifier = ">=0.2.2" },
     { name = "requests", specifier = ">=2.32.3" },
     { name = "tavily-python", specifier = ">=0.7.2" },
@@ -1835,6 +1864,21 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/8a/0b/9fcc47d19c48b59121088dd6da2488a49d5f72dacf8262e2790a1d2c7d15/pygments-2.19.1-py3-none-any.whl", hash = "sha256:9ea1544ad55cecf4b8242fab6dd35a93bbce657034b0611ee383099054ab6d8c", size = 1225293, upload-time = "2025-01-06T17:26:25.553Z" },
 ]
 
+[[package]]
+name = "pymupdf"
+version = "1.25.5"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/f9/af/3d5d363241b9a74470273cf1534436f13a0a61fc5ef6efd19e5afe9de812/pymupdf-1.25.5.tar.gz", hash = "sha256:5f96311cacd13254c905f6654a004a0a2025b71cabc04fda667f5472f72c15a0", size = 69812626, upload-time = "2025-03-31T23:22:13.891Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/85/5f/153d6c338291448e182648844849d13938a62a82a3e4a9b0907d9b381148/pymupdf-1.25.5-cp39-abi3-macosx_10_9_x86_64.whl", hash = "sha256:cde4e1c9cfb09c0e1e9c2b7f4b787dd6bb34a32cfe141a4675e24af7c0c25dd3", size = 19364722, upload-time = "2025-03-31T23:18:04.729Z" },
+    { url = "https://files.pythonhosted.org/packages/4e/55/43b64fa6cd048d2ea4574c045b5ac05d023254b91c2c703185f6f8a77b30/pymupdf-1.25.5-cp39-abi3-macosx_11_0_arm64.whl", hash = "sha256:5a35e2725fae0ab57f058dff77615c15eb5961eac50ba04f41ebc792cd8facad", size = 18606161, upload-time = "2025-03-31T23:18:28.707Z" },
+    { url = "https://files.pythonhosted.org/packages/8b/22/29edb3236aed2f99a7922699fd71183e2f6cdde3c3884670158ae4dcf3ea/pymupdf-1.25.5-cp39-abi3-manylinux2014_aarch64.manylinux_2_17_aarch64.whl", hash = "sha256:d94b800e9501929c42283d39bc241001dd87fdeea297b5cb40d5b5714534452f", size = 19467121, upload-time = "2025-04-01T09:31:13.172Z" },
+    { url = "https://files.pythonhosted.org/packages/18/12/95e2ebe2933f94800fdeafd87bc281a790e1dc947b147c3d101df4f73703/pymupdf-1.25.5-cp39-abi3-manylinux2014_x86_64.manylinux_2_17_x86_64.whl", hash = "sha256:ee22155d3a634642d76553204867d862ae1bdd9f7cf70c0797d8127ebee6bed5", size = 20030310, upload-time = "2025-03-31T23:19:03.607Z" },
+    { url = "https://files.pythonhosted.org/packages/bd/db/b4edec9e731ea7c2b74bf28b9091ed4e919d5c7f889ef86352b7fd416197/pymupdf-1.25.5-cp39-abi3-musllinux_1_2_x86_64.whl", hash = "sha256:6ed7fc25271004d6d3279c20a80cb2bb4cda3efa9f9088dcc07cd790eca0bc63", size = 21293562, upload-time = "2025-03-31T23:19:34.612Z" },
+    { url = "https://files.pythonhosted.org/packages/ec/47/682a8ddce650e09f5de6809c9bce926b2493a19b7f9537d80d4646989670/pymupdf-1.25.5-cp39-abi3-win32.whl", hash = "sha256:65e18ddb37fe8ec4edcdbebe9be3a8486b6a2f42609d0a142677e42f3a0614f8", size = 15110464, upload-time = "2025-03-31T23:20:02.136Z" },
+    { url = "https://files.pythonhosted.org/packages/71/c2/a9059607f80dcaf2392f991748cfc53456820392c0220cff02572653512a/pymupdf-1.25.5-cp39-abi3-win_amd64.whl", hash = "sha256:7f44bc3d03ea45b2f68c96464f96105e8c7908896f2fb5e8c04f1fb8dae7981e", size = 16579671, upload-time = "2025-03-31T23:20:25.793Z" },
+]
+
 [[package]]
 name = "python-dateutil"
 version = "2.9.0.post0"
@@ -2164,6 +2208,12 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/b1/93/dba5ed08c2e31ec7cdc2ce75705a484ef0be1a2fecac8a58272489349de8/setuptools-80.4.0-py3-none-any.whl", hash = "sha256:6cdc8cb9a7d590b237dbe4493614a9b75d0559b888047c1f67d49ba50fc3edb2", size = 1200812, upload-time = "2025-05-09T20:42:25.325Z" },
 ]
 
+[[package]]
+name = "sgmllib3k"
+version = "1.0.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/9e/bd/3704a8c3e0942d711c1299ebf7b9091930adae6675d7c8f476a7ce48653c/sgmllib3k-1.0.0.tar.gz", hash = "sha256:7868fb1c8bfa764c1ac563d3cf369c381d1325d36124933a726f29fcdaa812e9", size = 5750, upload-time = "2010-08-24T14:33:52.445Z" }
+
 [[package]]
 name = "shellingham"
 version = "1.5.4"