Among other things, I was in charge of the assurance documentation and testing. The system assumed an honest-but-curious threat model, where users may attempt to:
Bypass authorization rules
Brute-force authentication mechanisms
Tamper with the application or audit state
Intercept or manipulate network traffic
I made sure that the system protected against:
Credential compromise (via strong password policies and rate limiting)
Privilege escalation (via strict role enforcement)
Unauthorized access (via session-based authentication)
Audit log tampering (hash chaining with hmac verification)
Network eavesdropping (via TLS 1.3)