artifact-design
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATION
Full Analysis
- [SAFE]: The skill serves as a technical design manual for the AI agent to follow when creating web artifacts. It does not contain any executable code or malicious instructions.
- [EXTERNAL_DOWNLOADS]: Defines a restricted set of well-known and trusted external resources (such as Google Fonts and Cloudflare CDNs) that can be utilized within generated artifacts, aligning with security best practices for Content Security Policy (CSP) enforcement.
- [DATA_EXFILTRATION]: Explicitly instructs the agent to ensure that data stored in browser-side storage (localStorage, IndexedDB) remains private to the viewer and is not transmitted back to the AI or third parties.
- [SAFE]: No patterns of prompt injection, obfuscation, or unauthorized privilege escalation were identified. The instructions promote secure coding practices like version pinning and sanitization.
Audit Metadata