anti-ai-writing-patterns
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process arbitrary user-supplied text for rewriting, which is an inherent surface for indirect prompt injection.
- Ingestion points: The 'input text' identified in the Process section.
- Boundary markers: The skill lacks explicit instructions to use delimiters (like triple quotes or XML tags) to isolate user content from the agent's instructions.
- Capability inventory: The skill does not request or use any external tools, file system access, or network capabilities, which significantly mitigates the risk of an injection attack.
- Sanitization: No input filtering or sanitization steps are defined for the text being humanized.
- [SAFE]: The skill contains no executable scripts, remote dependencies, or sensitive data access. The external reference to Wikipedia's 'Signs of AI writing' is a well-known and reputable knowledge source.
Audit Metadata