xhs-static-cards
Pass
Audited by Gen Agent Trust Hub on Jun 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill automates the project lifecycle by executing "npm install", "npm run build", and "npm run export" within a user-specified project directory.
- [EXTERNAL_DOWNLOADS]: The workflow relies on several peer skills (such as "web-typography" and "frontend-design") installed via the "npx skills add" command. It also downloads the "playwright" library and the Chromium browser binary as part of the setup process.
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it renders user-provided text into HTML previews.
- Ingestion points: User input is parsed into "content.json" in the project directory as described in Step 3 of SKILL.md.
- Boundary markers: No specific boundary markers or instruction-ignoring directives are applied to the user data in the generated JSON or HTML.
- Capability inventory: The skill uses "templates/scripts/build.mjs" to write HTML files and "templates/scripts/export.mjs" to execute a headless browser for screenshot generation.
- Sanitization: The "templates/scripts/build.mjs" script implements an "esc()" function that sanitizes HTML special characters ("&", "<", ">", "") from user input before rendering, effectively preventing malicious script injection within the preview environment.
Audit Metadata