pen-design

Pass

Audited by Gen Agent Trust Hub on Nov 3, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions for the agent to download its own definition file (SKILL.md) from reputable content delivery networks such as unpkg.com and jsdelivr.net. This is a standard procedure for retrieving package-related documentation and maintaining version synchronization.
  • [COMMAND_EXECUTION]: The skill requires the installation and execution of the @pen.dev/cli tool. These operations are performed using the npm package manager and the pen command, which are necessary to generate the design files requested by the user.
  • [INDIRECT_PROMPT_INJECTION]: The skill defines an interface for processing user-supplied data which represents a common injection surface.
  • Ingestion points: User-provided design descriptions and reference files are passed as arguments to the CLI tool via the --prompt and --prompt-file flags.
  • Boundary markers: To mitigate risk, the skill explicitly instructs the agent to pass user prompts directly without adding its own details, reducing the chance of the agent accidentally executing instructions hidden within the user's design request.
  • Capability inventory: The skill's capabilities are limited to command execution for the design tool and writing resulting design/image files to the project directory.
  • Sanitization: The skill relies on the internal security and validation logic of the pen.dev CLI's design agent to handle the user-provided prompt content safely.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 3, 2026, 10:28 AM