sketch
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to include external resources in the generated HTML mockups. Specifically, it uses Tailwind CSS via the official CDN (https://cdn.tailwindcss.com) and Google Fonts via the standard tag. These are well-known and established services used for web development.
- [INDIRECT_PROMPT_INJECTION]: The skill contains an attack surface for indirect prompt injection because it ingests untrusted user input (design feel, references, and core actions) and incorporates them into the generation of executable HTML files and styling instructions.
- Ingestion points: User-provided strings for 'Feel', 'References', and 'Core action' collected during the intake phase described in SKILL.md.
- Boundary markers: The instructions do not specify explicit delimiters or "ignore instructions" markers for the user-provided design specifications.
- Capability inventory: The skill utilizes file writing (
write_file), terminal command execution (terminal), and browser-based rendering and analysis (browser_navigate,browser_vision) to process and verify the mockups. - Sanitization: There is no mention of sanitization, escaping, or validation of the user input before it is interpolated into the generated HTML files.
Audit Metadata