web-interface-architect
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [Prompt Injection] (SAFE): The text contains markers like 'CRITICAL' and 'IMPORTANT', but these are used for stylistic emphasis regarding design quality. There are no attempts to bypass safety filters, extract system prompts, or override core agent instructions.
- [Data Exposure & Exfiltration] (SAFE): No sensitive file paths, hardcoded credentials, or network operations (curl, wget, etc.) were detected.
- [Unverifiable Dependencies & Remote Code Execution] (SAFE): The skill does not reference or install external packages. It mentions libraries like 'Motion' as a suggestion for the user's generated code but does not execute or download them itself.
- [Dynamic Execution] (SAFE): There are no scripts or commands that involve dynamic code evaluation or runtime compilation.
- [No Code] (SAFE): The skill consists entirely of markdown instructions and metadata with no accompanying executable scripts or configuration files that could introduce technical vulnerabilities.
Audit Metadata