od-deck-ppt-keynote
Pass
Audited by Gen Agent Trust Hub on Oct 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill uses the
$ARGUMENTSvariable to capture user input and incorporates it directly into a brief for the Open Design MCP server. This brief then generates instructions that the agent is explicitly told to follow using its file-editing tools, creating a multi-step instruction chain. - Ingestion points: User input via the
$ARGUMENTSplaceholder inSKILL.md. - Boundary markers: Absent; user input is mixed directly into the instructional string.
- Capability inventory: The agent is instructed to use its file-editing tools and call
register_open_design_artifactbased on the output of the design tool. - Sanitization: No escaping or validation is performed on the user input before it is passed to the tool.
Audit Metadata