vercel-data-handling
Pass
Audited by Gen Agent Trust Hub on Mar 12, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface through its PII detection and data redaction functions.
- Ingestion points: Untrusted data is ingested from external sources via functions like
detectPII(text),redactPII(data), andexportUserData(userId)inreferences/implementation.md. - Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the provided templates for processing external data.
- Capability inventory: The skill allows
Read,Write, andEditoperations and includes logic for database deletions and Vercel API interactions. - Sanitization: The implementation provides redaction logic but lacks mechanisms to sanitize or escape data to prevent the execution of instructions contained within the text or objects being processed.
Audit Metadata