Hi cPFence
I’d like to suggest a feature that could significantly improve SPAM detection and overall user experience.
I often advise my clients to move unwanted emails to the SPAM folder when they identify messages that passed the filters but are clearly SPAM. It would be extremely useful if CPfence could scan the contents of the SPAM folder (via IMAP) one or more times per day, per email account, and perform one of the following actions:
Learn from those messages and mark similar ones as suspicious in the future;
Optionally, block messages from the same sender/domain or with the same content pattern;
Enable a “learning mode” that continuously improves filtering based on real user feedback.
This would create a user-based SPAM feedback loop, similar to what is implemented by some advanced anti-spam systems.
I believe this feature would make CPfence even more effective, especially in shared hosting environments where user input is key to reducing false negatives.
I’d be happy to hear your thoughts — and to know whether something like this is already on the roadmap.
Best regards,
Joao Costa
