marketing-performance-report
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill operates on data ingested from external sources (marketing platforms), creating an inherent attack surface for indirect instructions.
- Ingestion points: Performance metrics and GA4 data points provided by the user in SKILL.md.
- Boundary markers: Instructions do not define specific delimiters for separating user data from system instructions.
- Capability inventory: The skill's functionality is confined to text synthesis, data presentation, and routing to internal visualization tools, with no evidence of file system or network access.
- Sanitization: There are no explicit data sanitization or filtering protocols mentioned in the skill definition.
Audit Metadata