audit

Pass

Audited by Gen Agent Trust Hub on Jul 15, 2026

Risk Level: SAFE
Full Analysis
  • Systematic Audit Workflow: The skill establishes a structured process for capturing product flows, including specific instructions for loading browser tools, verifying screenshot integrity, and ensuring screenshots are visibly placed in Figma or local folders. This provides a clear audit trail for the agent's actions.
  • Evidence Validation: The instructions mandate multiple checks to ensure captured screenshots are not blank, loading, or incorrect before they are accepted as audit evidence. This prevents the agent from making conclusions based on invalid or missing data.
  • Defined Scope and Guardrails: The design audit framework limits the agent's analysis to user goals, interaction flows, and accessibility risks rather than business strategy. It also explicitly prevents the agent from claiming full WCAG compliance based solely on visual evidence.
  • Controlled Destination Handling: The skill uses explicit rules for output destinations (Figma or local storage) and requires the agent to confirm the user's preference if it is not stated, ensuring data is placed only where intended.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 15, 2026, 02:40 AM
Security Audit — agent-trust-hub — audit