sales-page-copywriting
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests product and market descriptions from the user to drive its generation process.
- Ingestion points: Product DNA and Market Context inputs in PHASE 1.
- Boundary markers: Absent; the instructions do not wrap user inputs in specific delimiters.
- Capability inventory: No command execution, file system access, or network capabilities are requested in the skill frontmatter or body.
- Sanitization: No explicit sanitization of user-provided business details is performed.
- [SAFE]: The output constraints (e.g., 'Do NOT show the user') are functional directives to the agent to manage its internal scratchpad and final output format for user experience, rather than malicious attempts to obfuscate behavior from safety filters.
Audit Metadata