cluster-agent-swarm-hardened
Warn
Audited by Socket on Apr 21, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS. The skill’s stated platform-operations purpose broadly matches its capabilities, and its guardrails argue against overt malicious intent. However, it explicitly performs transitive skill installation from an unpinned third-party GitHub repository, giving a broad multi-agent package significant operational reach over clusters and cloud tooling without strong provenance controls.
Confidence: 87%Severity: 74%
Audit Metadata