dependency-vulnerability-scan
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill uses the
$ARGUMENTSplaceholder to incorporate user-supplied input into the system prompt. This creates a potential surface for indirect prompt injection where malicious instructions could be embedded in the provided arguments. However, the skill lacks any defined tools or executable scripts, significantly limiting the risk. - Ingestion points:
SKILL.md(via$ARGUMENTSvariable in the Context section) - Boundary markers: None present (input is directly interpolated into the prose)
- Capability inventory: None (The skill contains no scripts, tools, or command execution logic)
- Sanitization: None present (The skill does not provide instructions for validating or escaping the input)
- [SAFE]: The skill consists of static documentation and best practice recommendations. All referenced external links point to reputable security organizations and established industry tools.
Audit Metadata