martech-stack
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill consists entirely of instructional markdown files without any executable code, scripts, or tool configurations.
- [INDIRECT_PROMPT_INJECTION]: The skill instructions address potential indirect prompt injection vulnerabilities by defining boundaries for data processing.
- Ingestion points: The skill describes analyzing text from exports, tickets, survey answers, and web pages (SKILL.md).
- Boundary markers: Explicit instructions state that such text must be treated as data and never as instructions to the agent (SKILL.md).
- Capability inventory: The skill provides no executable tools, scripts, or network capabilities.
- Sanitization: Instructions mandate using aggregates and the least personal data required for a task.
Audit Metadata