od-deck-deck-open-slide-canvas
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 content via the
$ARGUMENTSvariable to generate a design brief, which is then used by the agent to author files based on returned instructions. - Ingestion points: User input enters the system through the
$ARGUMENTSplaceholder inSKILL.mdand is passed to theprepare_open_design_brieftool. - Boundary markers: The instructions do not define explicit delimiters or 'ignore' instructions to wrap the user-supplied content.
- Capability inventory: The agent uses file-editing tools and the
register_open_design_artifacttool to perform actions based on the generated brief. - Sanitization: No explicit sanitization or validation of the input is specified before it influences the agent's file-editing actions.
Audit Metadata