pencil
Pass
Audited by Gen Agent Trust Hub on Mar 21, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill facilitates the ingestion of external data from
.pendesign files and guidelines, which represents a potential surface for indirect prompt injection where instructions could be embedded in design metadata. - Ingestion points:
batch_get,get_variables, andget_guidelinestools. - Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the prompt templates.
- Capability inventory: The skill has the ability to execute design scripts via
batch_design, write to files viaset_variables, and export visual data viaget_screenshot. - Sanitization: No sanitization or validation of the ingested design data is described.
- [COMMAND_EXECUTION]: The
batch_designtool uses a domain-specific language (DSL) to execute a sequence of operations provided as a string. This allows the agent to dynamically generate and execute logic within the Pencil environment.
Audit Metadata