Currently, the “Restrict Content Detection” option supports only text files, but when enabled, it generates too many false positives. This happens because EPP triggers on all text files, whereas we specifically need to monitor .CSV and .TXT files.
At the moment, the “Restrict Content Detection” option does support .CSV file types, but it does not function as we expect. Instead of simply checking files by extension (.csv), it also verifies the file format based on the specification described here. This means that if a user modifies the internal format of a CSV file, EPP may fail to recognize it as a CSV file, allowing potential data leaks.
To address this issue, we believe a better approach would be to allow “Restrict Content Detection” to check files based solely on their extensions. This way, the CAP policy would scan only files with the specified extensions, ensuring more precise content detection.
Similarly, we suggest this approach should also apply to the .TXT extension, which is currently missing from the available file types in the “Restrict Content Detection” option.
Implementing this change would significantly enhance EPP’s accuracy and usability while preventing unnecessary false positives.
Would love to hear feedback from the community on this! 