// AI Platform Drift

Cloudflare One adds Periodic Beaconing Traffic detection to Gateway

Cloudflare developer changelog · published 2026-11-16 22:01 UTC
WORTH KNOWING VERIFY_ONLY SECURITY
WHAT CHANGEDCloudflare One Gateway now detects periodic beaconing traffic as a risk behavior. The system flags a user after more than 30 contacts to the same destination in six hours, with contacts evenly spaced from 10 seconds to about 12 minutes apart. Detection is based on traffic behavior only, not threat intelligence or reputation. The feature is off by default with Info risk level and is available on Enterprise plans only.
WHY IT MATTERSPeriodic beaconing is a common command-and-control pattern, but legitimate software also checks in regularly. Each detection is a lead to investigate, not proof of compromise. Without enabling this behavior in Risk score settings, your organization will not receive these alerts even if beaconing traffic is present.
WHAT TO DOIf you are on an Enterprise plan, go to Risk score > Risk behaviors in Cloudflare One and decide whether to enable Periodic Beaconing Traffic detection. Events arrive 3 to 6 hours after the six-hour detection window closes.
Model-reported confidence 85% (not measured accuracy) Primary source ↗ written by anthropic/claude-sonnet-4.5
Show the reasoning — raw diff, triage verdict, confidence
RAW DIFF
+## Cloudflare One, Gateway - Cloudflare One adds Periodic Beaconing Traffic detection
+https://developers.cloudflare.com/changelog/post/2026-11-16-periodic-beaconing-traffic-ueba/
+Mon, 16 Nov 2026 00:00:00 GMT
+
+Cloudflare One now includes Periodic Beaconing Traffic as a risk behavior. It analyzes Gateway HTTP traffic for regular check-ins to the same destination.
+Cloudflare flags a user after more than 30 contacts to the same destination in six hours. Contacts must be evenly spaced, from 10 seconds to about 12 minutes apart. Few other users in your organization contact that destination.
+Detection uses traffic behavior only. It does not use threat intelligence, categories, or reputation. Malware can check in with a command-and-control (C2) server this way, but legitimate software also checks in regularly. Treat each event as a lead to investigate, not proof that a destination is C2.
+By default, the behavior is off and its risk level is Info. To turn it on, go to Risk score > Risk behaviors. Cloudflare reports a continuing beacon once. Events usually arrive 3 to 6 hours after the six-hour window closes.
+Periodic Beaconing Traffic is available on Enterprise plans.
HOW IT WAS JUDGED
Triage signals below are model-reported assertions, preserved verbatim; they are not independent arithmetic verification.
triage verdictMATERIAL at 0.92, needed 0.5
triage reasoningNew security detection capability added to Cloudflare One Gateway with plan-level gating and operational thresholds.
signalsnew feature addition to security product · feature availability gated to Enterprise plans · new detection behavior with specific thresholds (30 contacts in 6 hours) · new risk behavior category added to Risk score system · deployment/configuration requirement (must be enabled)
suppression—
triage modelanthropic/claude-haiku-4.5 ($0.00000)
writer modelanthropic/claude-sonnet-4.5 ($0.00000)