agent-sessions
Warn
Audited by Socket on May 8, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS: the skill’s purpose and file access are mostly coherent, and no clear exfiltration or credential harvesting is shown. However, it depends on an unverifiable local script as its primary mechanism and also references third-party history tools, so execution trust is too weak to rate benign.
Confidence: 87%Severity: 78%
Audit Metadata