pencil-mcp-batch-design
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided content (e.g., text content and component properties) for inclusion in design nodes via the
batch_designtool, creating a vulnerability surface. - Ingestion points: The
operationsparameter in thebatch_designtool call (defined inSKILL.md) accepts arbitrary properties and node data. - Boundary markers: The skill instructions do not specify any delimiters or warnings to ignore instructions embedded within the design data.
- Capability inventory: The skill uses the
batch_designtool to modify canvas elements (SKILL.md). - Sanitization: The skill lacks instructions for sanitizing or validating user-provided content before it is passed to the design tool.
Audit Metadata