humanizer-he

Pass

Audited by Gen Agent Trust Hub on Jul 23, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides detailed linguistic instructions for editing Modern Hebrew text, focusing on morphology, register, and cultural idiom.
  • [SAFE]: No malicious patterns, such as prompt injection, data exfiltration, or privilege escalation, were detected.
  • [SAFE]: The external reference provided in the metadata points to a legitimate public repository on GitHub, which is a well-known service.
  • [SAFE]: The skill uses standard platform tools (Read, Write, Edit, AskUserQuestion) for text processing and user interaction without any dangerous command execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 23, 2026, 05:06 PM
Security Audit — agent-trust-hub — humanizer-he