hunt-agent-prompt-extraction
Warn
Audited by Socket on Nov 2, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
The skill is a purpose-built prompt-extraction and agent-reconnaissance guide. It does not show credential exfiltration to a fixed third party or an unsafe install chain, but it intentionally teaches bypassing protections to recover hidden instructions and potentially actionable secrets. High security risk; malicious intent is not confirmed.
Confidence: 96%Severity: 78%
Audit Metadata