
Embedded AI

Sensor Operations
• Problem: Networks often harbor unpatched services and hidden CVEs across common ports (21, 22, 23, 80, 443, 3306, 3389).
• Solution: PentestSensor scans subnets, fingerprints service banners, cross -references a CVE SQLite database, and raises alerts on vulnerable endpoints.
• Why Different: Embedded ONNX AI models with adaptive thresholds, autonomous learning, and CVE spike heuristics deliver real -time anomaly detection and proactive defense.
• Syslog Output Fields: Timestamp, Sitename, Host, Subnet, IP, Port, Product, Version, CVE ID, Description, Vulnerability Count, Threat Level, Threat Score, Recommended Action.
Host Resource Requirements
-
CPU: < 1%
-
Memory(RAM) < 50MB
-
Disk: < 1MB per day log writes
-
Network: < 1KB/s syslog traffic




