penpot-uiux-design

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions to clone the penpot-mcp repository from the official Penpot GitHub organization (github.com/penpot/penpot-mcp.git). This is an official source required for the skill's primary functionality.
  • [COMMAND_EXECUTION]: The skill utilizes the mcp__penpot__execute_code tool to run JavaScript within the Penpot plugin context. This is a design-time capability used to automate the creation and modification of UI elements within a design file.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from existing design files (e.g., using penpotUtils.shapeStructure() to analyze hierarchies). While this constitutes an untrusted data ingestion surface, the risk is localized to the design environment and mitigated by standard agent behavior.
  • Ingestion points: Penpot shape data and layer structures read via mcp__penpot__execute_code (SKILL.md).
  • Boundary markers: None explicitly defined for design data ingestion.
  • Capability inventory: JavaScript execution in plugin context, image importation, and shape exportation (SKILL.md).
  • Sanitization: Not present for ingested design properties.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 05:33 PM
Security Audit — agent-trust-hub — penpot-uiux-design