ui-ux-pro-max

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOWPROMPT_INJECTION
Full Analysis
  • [Indirect Prompt Injection] (LOW): The tool processes and displays content from external CSV files, which serves as an ingestion point for untrusted data.\n
  • Ingestion points: Multiple CSV files located in the data/ directory (e.g., ux-guidelines.csv, prompts.csv) are loaded via _load_csv in scripts/core.py.\n
  • Boundary markers: No delimiters or isolation instructions are used to wrap search results, meaning any embedded instructions in the data could be interpreted by the LLM as higher-priority commands.\n
  • Capability inventory: The skill is strictly limited to data retrieval and text output; it lacks dangerous capabilities such as network access, subprocess execution, or system file writing.\n
  • Sanitization: CSV content is passed directly to the output without any escaping or filtering of potentially malicious prompt strings.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 09:29 AM