CSV MCP Server

Instructions

Get statistical summary of numeric columns in the CSV file.

Args:
    filename: Name of the CSV file

Returns:
    Dictionary with statistical analysis of numeric columns

Implementation Reference

• csv_mcp_server/server.py:163-177 (handler)

  MCP tool handler and registration for 'get_statistics'. This decorated function handles the tool invocation and delegates to the CSVManager implementation.

  @mcp.tool()
  def get_statistics(filename: str) -> Dict[str, Any]:
      """
      Get statistical summary of numeric columns in the CSV file.
      
      Args:
          filename: Name of the CSV file
      
      Returns:
          Dictionary with statistical analysis of numeric columns
      """
      try:
          return csv_manager.get_statistics(filename)
      except Exception as e:
          return {"success": False, "error": str(e)}

• csv_mcp_server/csv_manager.py:320-358 (helper)

  Core implementation of get_statistics method in CSVManager class. Loads the CSV, selects numeric columns, computes descriptive statistics using pandas.describe(), and returns formatted results.

  def get_statistics(self, filename: str) -> Dict[str, Any]:
      """Get statistical summary of numeric columns in the CSV file."""
      filepath = self._get_file_path(filename)
      
      if not filepath.exists():
          raise FileNotFoundError(f"CSV file '{filename}' not found")
      
      try:
          df = pd.read_csv(filepath)
          
          # Get numeric columns only
          numeric_df = df.select_dtypes(include=['number'])
          
          if numeric_df.empty:
              return {
                  "success": True,
                  "filename": filename,
                  "message": "No numeric columns found",
                  "statistics": {}
              }
          
          # Convert describe results to serializable format
          stats_dict = {}
          for col in numeric_df.columns:
              col_stats = numeric_df[col].describe()
              stats_dict[col] = {stat: float(value) if pd.notna(value) else None 
                               for stat, value in col_stats.items()}
          
          return {
              "success": True,
              "filename": filename,
              "statistics": stats_dict,
              "numeric_columns": list(numeric_df.columns),
              "total_columns": len(df.columns),
              "null_counts": {col: int(count) for col, count in numeric_df.isnull().sum().items()}
          }
      except Exception as e:
          logger.error(f"Failed to get statistics: {e}")
          raise