ralph
Warn
Audited by Socket on Apr 19, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS: the skill's planning/specification purpose is partly coherent, but its footprint is broader than necessary. Persistent autonomous looping, parallel sub-agents, shell access, web research with write/exec capability, and especially transitive installation from a different GitHub org make the overall skill higher risk than a normal methodology guide.
Confidence: 82%Severity: 72%
Audit Metadata