protocol-participation
Warn
Audited by Socket on Oct 2, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS. The skill is internally coherent for a network-control purpose, and its package/data-flow footprint is mostly proportionate. However, it grants an AI agent the ability to make live routing changes with root/network privileges, while relying on documented guardrails rather than enforceable technical controls; that makes it high operational risk even without signs of credential theft or malware.
Confidence: 87%Severity: 72%
Audit Metadata