security-policy-template
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides references to official and well-known security standard organizations, including NIST, ISO, SANS, and CIS, which are established and trusted sources for policy frameworks.\n- [INDIRECT_PROMPT_INJECTION]: The skill features an ingestion surface for untrusted data through the $ARGUMENTS placeholder.\n
- Ingestion points: The $ARGUMENTS variable is interpolated directly into the system context in SKILL.md.\n
- Boundary markers: No delimiters or instructions are used to distinguish the injected variable from the skill's instructions.\n
- Capability inventory: None. The skill does not use any tools, shell commands, file writing operations, or network requests.\n
- Sanitization: No validation or sanitization of the input is performed.\n
- Note: The lack of exploitable capabilities ensures that the risk remains at a safe level, as the agent cannot perform actions outside of text generation.
Audit Metadata