writing-ad-copy
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted data from the user to generate ad copy.
- Ingestion points: The skill gathers "offer and audience details" from the user as part of its primary workflow in
SKILL.md. - Boundary markers: Absent; there are no instructions to use delimiters or to ignore embedded instructions within the user-provided context.
- Capability inventory: The skill is limited to text generation; it does not request any tools, shell access, or network capabilities in its frontmatter.
- Sanitization: Absent; the instructions do not include steps to filter or escape the user-provided inputs before processing them into templates.
Audit Metadata