modern-web-app-architecture
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- Prompt Injection (SAFE): No evidence of jailbreak attempts or instruction overrides in the skill's logic or metadata. The language is purely instructional and professional.\n- Data Exposure & Exfiltration (SAFE): No sensitive file paths, hardcoded secrets, or suspicious network operations were identified. The security reference file specifically advises against storing secrets in the browser.\n- Unverifiable Dependencies (SAFE): While the skill mentions various development tools (Vite, Playwright, Vitest), it does not attempt to install or execute them on the host system.\n- Obfuscation (SAFE): No use of Base64, zero-width characters, or other encoding techniques to hide malicious intent.\n- Dynamic Execution (SAFE): Code snippets provided in the 'design-patterns' and 'rendering-strategies' files are static educational examples and are not used for runtime code generation or execution.
Audit Metadata