frontend-ui-ux-wizard

Pass

Audited by Gen Agent Trust Hub on Aug 21, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill manages the installation of development tools like Playwright and official hosting CLIs (Vercel, Netlify, Cloudflare). It explicitly mandates sourcing these from verified official registries.
  • [DATA_EXFILTRATION]: Step 6 contains proactive security measures to prevent credential leaks, specifically instructing the agent not to commit node_modules, build outputs, or secrets/.env files to GitHub repositories.
  • [COMMAND_EXECUTION]: The skill executes standard production build and deployment commands (e.g., npm run build, CLI deploy commands), which are appropriate for its functional purpose.
  • [PROMPT_INJECTION]: The 'Trust boundaries' section addresses potential indirect prompt injection by instructing the agent to treat fetched documentation as reference only and to watch for malicious embedded directives. Mandatory Evidence Chain: 1. Ingestion points: User-provided branding, mood descriptions, and external framework documentation. 2. Boundary markers: Explicit 'Trust boundaries' section instructing the agent to ignore instructions in fetched content. 3. Capability inventory: Execution of npm, git, and hosting CLIs. 4. Sanitization: Relies on workflow constraints rather than explicit input filtering.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 21, 2026, 12:17 PM