auto-skill
Warn
Audited by Socket on Apr 8, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS. The skill’s main function is presented as memory management, but its actual footprint includes self-prioritization, background operation, and persistent modification of global agent rule files across environments. There is no clear exfiltration or malware payload, yet the self-installing prompt persistence and unprompted config edits are not proportionate to the stated purpose.
Confidence: 92%Severity: 78%
Audit Metadata