ai-humanizer
Pass
Audited by Gen Agent Trust Hub on Feb 13, 2026
Risk Level: LOWNO_CODEPROMPT_INJECTION
Full Analysis
- [NO_CODE] (SAFE): The skill consists of natural language instructions and contains no scripts or binaries.
- [PROMPT_INJECTION] (INFO): Category 8 (Indirect Prompt Injection) surface identified. 1. Ingestion: User-provided text to be rewritten. 2. Boundaries: Absent. 3. Capabilities: None (no network, shell, or file writes). 4. Sanitization: Absent. The risk is minimal as the skill lacks capabilities to act on malicious instructions.
- [EXTERNAL_DOWNLOADS] (INFO): Documentation includes links to external marketing and installation pages; no runtime downloads occur.
Audit Metadata