NYC

ux-researcher-designer

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOWPROMPT_INJECTION
Full Analysis
  • Indirect Prompt Injection (LOW): The skill is designed to ingest and analyze user data and interview transcripts through persona_generator.py.
  • Ingestion points: User data and interviews provided via JSON input.
  • Boundary markers: None mentioned in the skill description.
  • Capability inventory: The skill generates personas, journey maps, and research synthesis; it lacks network access or file-write capabilities in its description.
  • Sanitization: No sanitization or filtering of external interview content is described.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 06:46 AM