I tried to implement your idea on a toy dataset as it seems correct:

master_file_compare = pd.DataFrame({"SKU": range(10)})
account_data = pd.DataFrame({"SKU": range(5, 15)})

# Merging both dataframe using left join
comparison_result = pd.merge(master_file_compare,account_data, on='SKU', how='left', indicator=True)

# Filtering only the rows that are available in left (master_file_compare)
comparison_result = comparison_result.loc[comparison_result['_merge'] == 'left_only']
display(comparison_result)

    SKU _merge
0   0   left_only
1   1   left_only
2   2   left_only
3   3   left_only
4   4   left_only

It is not the result that you are looking for ?
In your code, instead of account-data.csv it shouldn't be current-data.csv ?