pixel-perfect-ui

Fail

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: HIGHPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [Indirect Prompt Injection] (HIGH): The skill implements a high-privilege workflow where external content from Figma is used to generate React components.
  • Ingestion points: scripts/extract-component.py processes design_data intended to be fetched from the Figma API (referenced in references/checklist-template.md via mcp0_get_design_context).
  • Boundary markers: None. The script uses direct string interpolation to place external design values into React templates.
  • Capability inventory: The script performs file-write operations to save generated .tsx or .jsx files to the local disk (scripts/extract-component.py).
  • Sanitization: Absent. There is no validation or escaping of the design data before it is written into the component code, allowing a malicious design source to inject arbitrary logic into the user's codebase.
  • [Command Execution] (MEDIUM): The Python scripts perform significant filesystem operations, including directory creation and writing files that are intended to be part of an active application's source code. This represents a risk if the generation process is subverted.
  • [External Downloads] (LOW): The scripts/visual-compare.py script relies on numpy and Pillow (PIL). While these are standard libraries for image processing, they are external dependencies that are not included in the standard Python library.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 16, 2026, 07:49 AM