symmetric-cipher-attacks

Pass

Audited by Gen Agent Trust Hub on Nov 9, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides legitimate Python and SageMath implementations for well-known cryptographic attacks (Padding Oracle, CBC, ECB, Stream Ciphers, MT recovery). These are standard educational resources for security research.
  • [SAFE]: No evidence of prompt injection or attempts to bypass safety filters were found. The instructional language is professional and scope-appropriate.
  • [SAFE]: The skill does not contain hardcoded credentials, sensitive file access, or unauthorized network exfiltration patterns. External tool references (PadBuster, xortool) are standard security utilities.
  • [SAFE]: No obfuscation techniques, hidden characters, or homoglyph attacks were detected in the source code or documentation.
  • [SAFE]: The skill does not employ persistence mechanisms, privilege escalation techniques, or dynamic context injection via shell command substitution.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 9, 2026, 11:59 AM