docker
Warn
Audited by Socket on Mar 23, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS. The core Docker-generation behavior is coherent and mostly uses standard local tooling, but the skill is overly agentic: it mandates autonomous execution, can run validation/build commands, writes telemetry into Claude memory directories, and instructs the agent to invoke additional unspecified skills. Those scope-expanding behaviors are disproportionate enough to raise medium-high security risk, though there is no clear evidence of credential theft or overt malware.
Confidence: 88%Severity: 71%
Audit Metadata