cto-review
Warn
Audited by Socket on Apr 26, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS: the skill’s GitHub review capabilities largely fit its stated purpose, and its main external CLI (gh) appears official. But it is high-impact because it reads raw local tokens, routes data to unverifiable localhost services, and grants the agent autonomous write/merge authority on GitHub; that footprint is disproportionate for a checklist-style review skill unless tightly sandboxed and explicitly approved.
Confidence: 89%Severity: 79%
Audit Metadata