od-video-vfx-text-cursor
Pass
Audited by Gen Agent Trust Hub on Oct 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill incorporates user-supplied input through the
$ARGUMENTSvariable into a design brief. - Ingestion points: The
$ARGUMENTSplaceholder inSKILL.mdallows arbitrary user text to be passed into the workflow. - Boundary markers: No explicit delimiters or instructions are used to separate user input from the rest of the brief, increasing the risk that the agent might interpret user text as direct instructions.
- Capability inventory: The skill utilizes MCP tools (
prepare_open_design_brief,register_open_design_artifact) and native file-editing tools to generate code and assets. - Sanitization: There is no evidence of input validation or sanitization performed on the
$ARGUMENTScontent before it is processed.
Audit Metadata