pencil

Pass

Audited by Gen Agent Trust Hub on Apr 6, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external .pen design files, creating a potential surface for indirect prompt injection if those files contain malicious instructions.
  • Ingestion points: Data is ingested from local files via tools such as batch_get, get_editor_state, and get_variables (File: SKILL.md).
  • Boundary markers: No explicit delimiters or instructions are provided to the agent to differentiate between design data and potential embedded commands.
  • Capability inventory: The skill can execute design scripts (batch_design), modify file variables (set_variables), and capture visual output (get_screenshot) (File: SKILL.md).
  • Sanitization: The skill does not describe any validation or sanitization of the content retrieved from the design files before it is processed by the agent.
  • [COMMAND_EXECUTION]: The batch_design tool uses a domain-specific language (DSL) described as 'operation scripts' to perform design changes. While this involves interpreting a series of commands, it appears restricted to the functionality of the Pencil design tool and does not indicate arbitrary shell execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 6, 2026, 10:44 AM