php-modern-standards

Pass

Audited by Gen Agent Trust Hub on Feb 28, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: No malicious override or bypass instructions were detected. The instruction to use the 'Superpowers plugin' is a behavioral directive for the agent's operational workflow rather than an attempt to bypass safety guidelines.
  • [DATA_EXFILTRATION]: No evidence of data exposure or exfiltration. There are no hardcoded credentials, sensitive file path accesses, or unauthorized network operations.
  • [REMOTE_CODE_EXECUTION]: No remote code execution patterns or unverifiable external dependencies were found. The skill consists entirely of static Markdown and PHP code examples.
  • [COMMAND_EXECUTION]: The skill explicitly warns against command injection and provides safe alternatives (e.g., using PHP's native filesize() instead of shell_exec('ls -lh')).
  • [OBFUSCATION]: No forms of obfuscation (Base64, zero-width characters, homoglyphs, or encoded strings) were detected in the source files.
  • [SAFE]: The content is highly defensive, offering detailed implementations for SQL injection prevention (using PDO prepared statements), XSS protection (output escaping), and secure password hashing (Argon2id).
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 28, 2026, 12:09 PM