od-video-frame-macos-notification
Pass
Audited by Gen Agent Trust Hub on Oct 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied input via the
$ARGUMENTSvariable and interpolates it directly into a brief for theprepare_open_design_brieftool. This creates a surface where a user could provide instructions aimed at influencing the design generation process. - Ingestion points: User input is ingested via the
$ARGUMENTSparameter in theSKILL.mdfile. - Boundary markers: The input is placed inside a string template without explicit delimiters or instructions for the agent to ignore embedded commands within that input.
- Capability inventory: The skill utilizes specialized MCP tools (
prepare_open_design_brief,register_open_design_artifact) and local file-editing capabilities to produce design files. - Sanitization: There is no evidence of input validation or sanitization before the data is passed to the design system tools.
Audit Metadata