guizang-page-skill
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill downloads font files from the official Google Fonts GitHub repository, which is a recognized trusted source.
- [COMMAND_EXECUTION]: Executes local Node.js scripts for brand theme processing and launches a headless browser via Puppeteer to generate PDF and image exports from the rendered HTML.
- [INDIRECT_PROMPT_INJECTION]: Ingests user-supplied Markdown documents to generate layout annotations and component upgrades; this ingestion of untrusted content represents a potential surface for indirect prompt injection, though it is mitigated by standard agent safety protocols.
- [DYNAMIC_EXECUTION]: Dynamically assembles HTML documents incorporating content generated by the AI (such as SVG and HTML fragments), which are subsequently loaded and processed within a browser context for document export tasks.
Audit Metadata