design-iterator

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is assessed as safe because its activities are limited to UI design improvement and research consistent with its stated purpose.
  • [COMMAND_EXECUTION]: The skill uses the 'agent-browser' tool for browser-based tasks like navigation and capturing screenshots, which are standard operations for its UI/UX analysis workflow.
  • [PROMPT_INJECTION]: A surface for indirect prompt injection exists during competitor research where external URLs are accessed via 'agent-browser'. Ingestion points: URL navigation in 'SKILL.md'. Boundary markers: No explicit delimiters or ignore instructions for external content are used. Capability inventory: browser interaction and local file modification. Sanitization: No filtering of retrieved web content is applied before analysis.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 08:04 PM