phantom-ai-coworker
Warn
Audited by Socket on Sep 12, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS. The skill’s broad capabilities generally match its stated purpose as an autonomous AI co-worker, so this is not obviously deceptive malware. However, its real footprint is extremely powerful: many credentials, host-level Docker control, shell execution, public endpoints, dynamic MCP tool creation, and autonomous outbound communications. The main concern is high operational/security risk and transitive trust expansion, not a hidden installer or clear credential-harvesting endpoint.
Confidence: 91%Severity: 88%
Audit Metadata