cart-recovery-sms
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill establishes an automation workflow that ingests customer-provided data—such as first names and product titles—directly into SMS templates. This represents a typical surface for indirect prompt injection, as malicious content within these fields could potentially influence the resulting message or agent behavior.
- Ingestion points: Customer identity and cart item data are ingested via the
AbandonedCartandCustomertypes defined insrc/types/cart.ts. - Boundary markers: The skill does not explicitly specify the use of delimiters or 'ignore' instructions when interpolating customer data into message templates.
- Capability inventory: The skill enables network-based SMS dispatching and automated job scheduling based on the processed data.
- Sanitization: The instructions emphasize legal compliance and delivery best practices (e.g., carrier filter avoidance) but do not mandate specific sanitization for interpolated data fields.
Audit Metadata