ads
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [SAFE]: The skill is strictly advisory and instructional, providing playbooks for performance marketing.
- [EXTERNAL_DOWNLOADS]: Reference files include standard implementation snippets for tracking pixels from well-known services (Google, Meta, LinkedIn, and TikTok). These are documented neutrally and intended for installation on the user's website, not for execution by the agent.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests context from local marketing files.
- Ingestion points: Reads .agents/product-marketing.md (referenced in SKILL.md).
- Boundary markers: Absent.
- Capability inventory: No execution capabilities (such as subprocess calls, network writes, or file writes) are present in the skill scripts.
- Sanitization: Absent.
- [NO_CODE]: The skill consists entirely of Markdown reference files and JSON evaluation tests, with no executable scripts or system commands included.
Audit Metadata