hunt-agent-prompt-injection
Warn
Audited by Socket on Nov 2, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
The skill is a coherent but high-risk agent red-teaming guide. It contains no hidden installer, credential collection, or fixed exfiltration destination; its main risk is enabling actionable prompt-injection, data-exfiltration, and unauthorized-access testing without an explicit authorization boundary.
Confidence: 94%Severity: 78%
Audit Metadata