landing-page-vercel
Pass
Audited by Gen Agent Trust Hub on Feb 19, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATION
Full Analysis
- PROMPT_INJECTION (LOW): Detected a surface for Indirect Prompt Injection (Category 8).
- Ingestion points: User input provided via command line arguments (
--name,--tagline,--features) or interactive mode as described inSKILL.mdphase 1. - Boundary markers: Absent. There are no instructions to delimit or ignore embedded instructions within the user-provided content during the scaffolding process.
- Capability inventory: The
scaffold.pyscript performs file system writes to generate the landing page files (index.html,script.js, etc.). - Sanitization: Absent. There is no evidence of sanitization or validation of user-provided strings before they are interpolated into the generated web page source code.
- EXTERNAL_DOWNLOADS (LOW): The skill instructs users to install the Vercel CLI (
npm i -g vercel). While Vercel is a well-known service and the organization is on the trusted list, the installation command triggers an external dependency download. - DATA_EXFILTRATION (LOW): The skill facilitates deployment to Vercel, which involves uploading generated files to a non-whitelisted domain (
vercel.com). This behavior is consistent with the primary purpose of the skill and does not involve sensitive file access.
Audit Metadata