static-web-artifacts-builder
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill utilizes local shell scripts for project management, avoiding all external network requests or remote code downloads. The implementation is transparent and restricted to the user's local workspace.\n- [SAFE]: The bundle script performs proactive security checks, scanning the generated HTML for external links to CDNs, fonts, or scripts, which enforces a self-contained security boundary for the artifacts.\n- [INDIRECT_PROMPT_INJECTION]: The skill has a vulnerability surface for indirect prompt injection as it transforms user-provided prompts into HTML and CSS code. \n
- Ingestion points: User prompts are used to generate and edit the content of the
index.htmlfile via the agent.\n - Boundary markers: The instructions do not define specific delimiters or "ignore" instructions to isolate user-controlled data from the HTML template structure.\n
- Capability inventory: The skill performs local file system operations, including directory creation (
mkdir) and file writing (cat,cp) through its bash scripts.\n - Sanitization: No explicit sanitization or escaping of user-provided input is implemented in the code generation workflow.
Audit Metadata