docker-privesc
Warn
Audited by Socket on Jul 31, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
The skill is internally coherent but highly dangerous: it is an offensive privilege-escalation playbook that enables an AI agent or operator to escape Docker containers and obtain host root. Install provenance is mostly official and there is no third-party credential forwarding, so this is not confirmed malware; however, the capability itself is an AI-agent exploit skill with destructive, persistence-enabling host actions, making overall security risk high.
Confidence: 95%Severity: 93%
Audit Metadata