design-url-to-code
Pass
Audited by Gen Agent Trust Hub on Jun 13, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill identifies an indirect prompt injection surface where data fetched from external websites could contain instructions intended to hijack the agent's behavior. It mitigates this risk by instructing the agent to treat fetched HTML/CSS as untrusted data and never allow it to redirect the primary task.
- [EXTERNAL_DOWNLOADS]: The skill's core functionality involves fetching content (HTML, CSS, assets) and taking screenshots of external URLs using web-fetch and browser automation tools.
- [COMMAND_EXECUTION]: The process of rebuilding the frontend locally involves executing commands to run development servers, perform type checking, and manage build tools (e.g., Vite, Tailwind CSS).
Audit Metadata