claude-style-coding

Pass

Audited by Gen Agent Trust Hub on Apr 29, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists of prompt-based instructions that guide the agent's behavior during software development tasks. It defines a 'Product Frame' and decision rules to improve UX outcomes.
  • [SAFE]: No sensitive data exposure or exfiltration patterns were identified. The skill does not access environment variables, credentials, or personal files.
  • [SAFE]: References to running validation scripts (such as ux_audit or accessibility_checker) describe a standard development workflow for UI/UX testing and do not involve executing untrusted remote code.
  • [SAFE]: The configuration file 'agents/openai.yaml' defines standard interface and policy settings for the skill without introducing any privilege escalation or persistence risks.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 29, 2026, 08:23 PM