ecommerce-copywriting
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external context and data files to generate marketing copy, which creates a potential surface for indirect prompt injection if those files are attacker-controlled. \n
- Ingestion points: The skill in
SKILL.mdexplicitly instructs the agent to read context from.claude/product-marketing-context.md,.claude/brand-guidelines.md, and product data files (CSV, JSON) for bulk descriptions. \n - Boundary markers: Absent; there are no instructions to the agent to treat external data as untrusted or to ignore embedded commands. \n
- Capability inventory: The skill consists entirely of Markdown guidelines and does not include any executable scripts, tool configurations, or network operations. \n
- Sanitization: Absent; no specific validation or sanitization steps are defined for the content extracted from the data files.
Audit Metadata