penpot-design-to-code-review
Pass
Audited by Gen Agent Trust Hub on Sep 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a strictly read-only workflow and explicitly forbids mutation of the design canvas within its logic. It includes mandatory checkpoints that require user confirmation before proceeding between analytical phases.
- [SAFE]: All interactions with the Penpot environment are performed via standard vendor-provided API methods (such as
penpot.generateMarkupandpenpot.selection) executed through the platform's authorized code execution tool. - [INDIRECT_PROMPT_INJECTION]: The skill has an ingestion surface for untrusted data, specifically when reading external source code or Storybook DOM structures for comparison. However, this data is processed strictly for property extraction and normalization (e.g., unit conversion, hex formatting) rather than instruction execution, making the risk negligible.
- [SAFE]: State management is handled through standard plugin storage mechanisms to maintain report data between audit steps. No attempts at unauthorized persistence or credential harvesting were detected.
Audit Metadata