customer-support
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists of instructional content defining a customer support persona. No malicious code, commands, or unauthorized data access patterns were found. The integration capabilities mentioned are conceptual and do not include risky implementation scripts.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to handle customer inquiries and data from external platforms (CRMs, chat tools). While the skill itself is safe, agents using these instructions will inherently have a surface for indirect prompt injection from customer-supplied content.
Audit Metadata