vercel-data-handling
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides comprehensive guidance on implementing privacy best practices, including PII redaction, secure cookie management, and data residency configurations. All code snippets are for implementation reference and trace back to the author's infrastructure.
- [INDIRECT_PROMPT_INJECTION]: The skill documents handling untrusted user data as part of its compliance functions. Ingestion points: Sensitive user data is ingested via Next.js API routes (api/users/[id]/route.ts) and middleware as described in SKILL.md. Boundary markers: The instructions explicitly define boundaries by teaching the use of redaction wrappers (redact()) before data is sent to logs or stored. Capability inventory: The implementation logic utilizes standard database operations and logging; no arbitrary command execution or unauthorized network access was found. Sanitization: The skill provides regex-based redaction patterns to identify and sanitize sensitive data types like emails, credit cards, and credentials.
Audit Metadata