ui-ux-pro-max

Pass

Audited by Gen Agent Trust Hub on Feb 20, 2026

Risk Level: SAFE
Full Analysis
  • SAFE (SAFE): The skill consists of static CSV data files and a Python script used to search that data. No malicious behavior or high-risk command execution patterns were found.
  • Indirect Prompt Injection (SAFE): The script processes user input as search queries to retrieve rows from local CSV files.
  • Ingestion points: User query via args.query in search.py.
  • Boundary markers: Absent in the output formatting, but retrieved content is constrained to the skill's own data files.
  • Capability inventory: No dangerous functions like subprocess, exec, or eval are present in the provided scripts.
  • Sanitization: Results are truncated and labeled, which provides a degree of context to the consuming agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 20, 2026, 08:03 PM