avoid-ai-writing
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied prose, which acts as an ingestion point for untrusted data. While the skill's instructions are purely editorial and it lacks dangerous capabilities like network access or file writing, it is theoretically susceptible to instructions hidden within the text being audited.
- Ingestion points: User-supplied prose text provided for auditing (SKILL.md).
- Boundary markers: None explicitly defined in the instruction set.
- Capability inventory: No command execution, file system access, or network operations detected in the skill instructions.
- Sanitization: No specific sanitization or escaping of the input text is described.
Audit Metadata