canvas-design
Pass
Audited by Gen Agent Trust Hub on Jun 15, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill's instructions are entirely consistent with its stated purpose of creating original visual designs and artistic manifestos. No malicious commands or unauthorized data access patterns were detected.\n- [EXTERNAL_DOWNLOADS]: The skill includes documentation and license files for a library of open-source fonts. It references official repositories and organizations such as Google Fonts, Vercel, and Red Hat, which are well-known and reputable sources for digital assets.\n- [PROMPT_INJECTION]: The instructions incorporate pre-seeded dialogue to guide the agent toward high-quality output (e.g., 'The user ALREADY said...'). This technique is used to enforce a specific creative persona and craftsmanship standard for design tasks, rather than to bypass safety protocols or override core agent behavior.
Audit Metadata