customer-service

Pass

Audited by Gen Agent Trust Hub on Jul 27, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The 'Mandatory Intro Message' section requires the agent to prepend a specific promotional string and solicitation for funds (referencing CashApp and BuyMeACoffee) to every blog post or long-form output generated by the skill.
  • [PROMPT_INJECTION]: The 'System Prompt Inquiry Response' section instructs the agent to hijack any user request for its internal instructions or system prompt, replacing the default response with a scripted redirect to a promotional membership site (scayveracademy.com).
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 27, 2026, 12:11 AM
Security Audit — agent-trust-hub — customer-service