LoFP / some benign applications may exhibit behaviors that resemble encrypted threat patterns, especially if they use uncommon encryption libraries or custom protocols. custom-developed or internal tools may trigger high eve confidence scores depending on how they encrypt data. it is recommended to validate the associated process (`eve_process`) and destination context, and correlate with other logs (e.g., endpoint or threat intel) before taking response action.

Techniques

• T1041
• T1071.001
• T1105
• T1573.002

Sample rules

Cisco Secure Firewall - High EVE Threat Confidence

• source: splunk
• technicques:
  • T1041
  • T1071.001
  • T1105
  • T1573.002

Description

The following analytic detects connections with a high Encrypted Visibility Engine (EVE) threat confidence score, indicating potentially malicious behavior within encrypted traffic. It leverages Cisco Secure Firewall Threat Defense logs and evaluates the EVE_ThreatConfidencePct field, which reflects the system’s confidence in classifying encrypted sessions as threats based on machine learning models and behavioral analysis. A score equal to or greater than 80 suggests the connection is highly likely to be associated with malware command and control (C2), remote access tools, or suspicious tunneling behavior. If confirmed malicious, this may indicate covert communication over TLS from compromised hosts.

Detection logic

`cisco_secure_firewall` EventType=ConnectionEvent EVE_ThreatConfidencePct >= 80

| stats count min(_time) as firstTime max(_time) as lastTime
    Values(rule) as rule
    Values(url) as url
    by EVE_Process, EVE_ThreatConfidencePct, src_ip, dest, dest_port, transport, action

| `security_content_ctime(firstTime)`

| `security_content_ctime(lastTime)`

| `cisco_secure_firewall___high_eve_threat_confidence_filter`