humanize
Pass
Audited by Gen Agent Trust Hub on Mar 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists of markdown and configuration files for text transformation. No executable scripts or binaries are included.
- [DATA_EXFILTRATION]: No evidence of sensitive data access or external transmission. The skill explicitly disclaims interaction with external APIs.
- [REMOTE_CODE_EXECUTION]: The skill does not perform any downloads or execute dynamic code from external sources.
- [PROMPT_INJECTION]: While the skill ingests untrusted text, it contains specific instructions to ignore embedded AI-typical instructions and focuses strictly on stylistic humanization with factual preservation checks.
Audit Metadata