animate
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data through the
$ARGUMENTSvariable to perform code modifications. - Ingestion points: The skill body in
SKILL.mdexplicitly instructs the agent to "Add or fix animations in$ARGUMENTS". - Boundary markers: There are no explicit delimiters (e.g., XML tags or triple quotes) or "ignore embedded instructions" warnings surrounding the interpolation of
$ARGUMENTS. - Capability inventory: The skill is authorized to perform direct file modifications ("edit code directly").
- Sanitization: The instructions do not describe any mechanisms for validating, filtering, or escaping the content provided in the arguments before it influences the agent's code-writing behavior.
Audit Metadata