Skip to main content
When you want to automate the handling of reported messages, you can use the Actions panel, although it gives just basic controls over what happens to the messages with malicious verdicts. For actions on a wider range of reported emails, you must resort to campaigns. Campaigns are saved search queries empowered with actions: message action (what to do with the email—quarantine, deliver, or don’t do anything) and resolve action (what final verdict should be applied—benign, malicious, or none). Using smartly crafted campaigns, you can automate the way XSP handles reported messages.
  • Always check results in your environment first before activating the automation: perform search queries to see if they produce expected results, and when they do, create campaigns.
  • Start with high-confidence spam and high-confidence phishing auto-resolution.
  • Run queries in the Abuse Mailbox / All Cases view (an equivalent of reported AND NOT(status:pending)) to ensure only reported messages are processed.
The following examples show some ideas on how to automate with campaigns. They might not work well in every environment. Simply copy and paste the provided search queries and check their results.
  • Resolve all reported messages with young domains (sender domain or embedded URL): (#365daysoldsender or #365daysold) and (#spam or #phishing)
  • Resolve all reported messages with shortened URLs (all confidences): #phishing #shortened
  • Resolve all reported internal emails without attachments: #internal not #nolocalreputation not #document
  • Resolve all reported partner emails without attachments or links to cloud storage: #partner #auth #trusted not #nolocalreputation not #cloudstorage not #document
  • Resolve all graymail without attachments or links to cloud storage: (#marketing or #crm or #newsletter or #bulk or #external) not #nolocalreputation not(#cloudstorage or #cloudstoragesender) not #document not #spffail not #dkimfail not #subjectcvip not #bodylt75words not #lowtrusted not #internal
Hints:
  • Use not #nolocalreputation not(#cloudstorage or #cloudstoragesender) to avoid any phishing false negatives.
  • Use not #document to avoid any malware false negatives.
Please refer to Tags to learn more about the tags used here.