Additional fixes to work with files and excel files.

agent_config.py

@@ -2,6 +2,10 @@ from smolagents import CodeAgent, DuckDuckGoSearchTool, HfApiModel, VisitWebpage
 from tools.wiki_tool import WikiTool
 from tools.excel_analysis_tool import ExcelAnalysisTool
 import os
+import requests
+import re
+import tempfile
+from pathlib import Path
 
 @tool
 def check_answer(answer: str) -> str:
@@ -29,6 +33,40 @@ def reverse_sentence_tool(reverse_sentence: str) -> str:
 
     return " ".join(correct_words)
 
+@tool
+def download_file_if_any(base_api_url: str, task_id: str) -> str | None:
+    """
+    Try GET /files/{task_id}.
+    • On HTTP 200 → save to a temp dir and return local path.
+    • On 404 → return None.
+    • On other errors → raise so caller can log / handle.
+    """
+    url = f"{base_api_url}/files/{task_id}"
+    try:
+        resp = requests.get(url, timeout=30)
+        if resp.status_code == 404:
+            return None          # no file
+        resp.raise_for_status()   # raise on 4xx/5xx ≠ 404
+    except requests.exceptions.HTTPError as e:
+        # propagate non-404 errors (403, 500, …)
+        raise e
+
+    # ▸ Save bytes to a named file inside the system temp dir
+    #    Try to keep original extension from Content-Disposition if present.
+    cdisp = resp.headers.get("content-disposition", "")
+    filename = task_id                                 # default base name
+    if "filename=" in cdisp:
+        m = re.search(r'filename="([^"]+)"', cdisp)
+        if m:
+            filename = m.group(1)                      # keep provided name
+
+    tmp_dir = Path(tempfile.gettempdir()) / "gaia_files"
+    tmp_dir.mkdir(exist_ok=True)
+    file_path = tmp_dir / filename
+    with open(file_path, "wb") as f:
+        f.write(resp.content)
+    return str(file_path)
+
 def create_agent():
     """
     Creates and configures the CodeAgent with the necessary model and tools.
@@ -60,11 +98,11 @@ def create_agent():
     # Create and return the agent
     agent = CodeAgent(
         model=model,
-        tools=[wiki_tool, web_search_tool, excel_analysis_tool, visitWebpageTool, python_interpreter_tool, check_answer, reverse_sentence_tool],
+        tools=[wiki_tool, web_search_tool, excel_analysis_tool, visitWebpageTool, python_interpreter_tool, check_answer, reverse_sentence_tool, download_file_if_any],
         add_base_tools=True,
         max_steps=8,
         verbosity_level=2,
-        additional_authorized_imports=['pandas','numpy','csv','subprocess']
+        additional_authorized_imports=['numpy','subprocess', 're', 'pandas', 'requests', 'json', 'os', 'pathlib', 'tempfile'],
     )
     return agent
 

requirements.txt

@@ -4,4 +4,6 @@ wikipedia
 pandas
 openpyxl
 huggingface_hub
-smolagents[litellm]
+smolagents[litellm]
+tabulate
+requests

tools/excel_analysis_tool.py

@@ -1,67 +1,63 @@
 from smolagents.tools import Tool
 import pandas as pd
+from tabulate import tabulate
+from typing import Union, Optional
 from huggingface_hub import hf_hub_download
 
 class ExcelAnalysisTool(Tool):
-    def __init__(self):
-        self.name = "ExcelAnalysisTool"
-        self.description = (
-            "Loads an Excel file from the GAIA dataset on Hugging Face and calculates "
-            "the total sales for items labeled as 'food', excluding drinks. "
-            "Provide input as a string with the filename, e.g., 'sales_data.xlsx'."
-        )
-        self.inputs = {
-            "filename": {
-                "type": "string",
-                "description": "The name of the Excel file to process (e.g., 'sales_data.xlsx').",
-                "nullable": True  # Сделать filename необязательным
-            }
-        }
-        self.output_type = "string"
-        self.repo_id = "gaia-benchmark/GAIA"
-        super().__init__(
-            name=self.name,
-            description=self.description,
-            inputs=self.inputs,
-            output_type=self.output_type
-        )
-
-    def forward(self, filename: str = "") -> str:
-        """
-        Loads and processes the Excel file.
-
-        Args:
-            filename (str): The name of the Excel file (e.g., 'sales_data.xlsx').
-
-        Returns:
-            str: Total food sales in USD, or an error message.
-        """
-        if not filename:
-            return "Error: 'filename' is required."
+    """Render an Excel worksheet as Markdown text."""
+
+    name = "excel_to_text"
+    description = (
+        "Read an Excel file and return a Markdown table of the requested sheet. "
+        "Accepts either the sheet name or the zero-based index."
+    )
+
+    inputs = {
+        "excel_path": {
+            "type": "string",
+            "description": "Path to the Excel file (.xlsx / .xls).",
+        },
+        "sheet_name": {
+            "type": "string",
+            "description": (
+                "Worksheet name or zero‑based index *as a string* (optional; default first sheet)."
+            ),
+            "nullable": True,
+        },
+    }
+
+    output_type = "string"
+
+    def forward(
+            self,
+            excel_path: str,
+            sheet_name: Optional[str] = None,
+    ) -> str:
+        """Load *excel_path* and return the sheet as a Markdown table."""
+
+        path = pathlib.Path(excel_path).expanduser().resolve()
+        if not path.exists():
+            return f"Error: Excel file not found at {path}"
 
         try:
-            # Download the file from Hugging Face Hub
-            file_path = hf_hub_download(
-                repo_id=self.repo_id,
-                filename=filename,
-                repo_type="dataset"
-            )
-
-            # Load the Excel file into a DataFrame
-            df = pd.read_excel(file_path)
-
-            # Filter rows: category == 'food' and item != 'drinks'
-            food_sales = df[
-                (df['category'].str.lower() == 'food') &
-                (df['item'].str.lower() != 'drinks')
-            ]
-
-            total_sales = food_sales['sales'].sum()
-            return f"Total sales for food items: ${total_sales:.2f}"
-
-        except FileNotFoundError:
-            return "Error: The specified file was not found."
-        except KeyError as e:
-            return f"Error: Missing expected column in the Excel file: {str(e)}"
-        except Exception as e:
-            return f"An unexpected error occurred: {str(e)}"
+            # Interpret sheet identifier -----------------------------------
+            sheet: Union[str, int]
+            if sheet_name is None or sheet_name == "":
+                sheet = 0  # first sheet
+            else:
+                # If the user passed a numeric string (e.g. "1"), cast to int
+                sheet = int(sheet_name) if sheet_name.isdigit() else sheet_name
+
+            # Load worksheet ----------------------------------------------
+            df = pd.read_excel(path, sheet_name=sheet)
+
+            # Render to Markdown; fall back to tabulate if needed ---------
+            if hasattr(pd.DataFrame, "to_markdown"):
+                return df.to_markdown(index=False)
+            from tabulate import tabulate  # pragma: no cover – fallback path
+
+            return tabulate(df, headers="keys", tablefmt="github", showindex=False)
+
+        except Exception as exc:  # broad catch keeps the agent chat‑friendly
+            return f"Error reading Excel file: {exc}"