supabase-pentest-skills
Warn
Audited by Socket on Sep 18, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS. The skill is internally consistent with its stated purpose, but that purpose is an offensive AI pentesting toolkit that automates credential extraction, unauthorized-read testing, account creation, and broad active probing. No clear credential exfiltration to unrelated domains is shown, but the combination of offensive capability, reduced confirmation guidance, transitive skill installation, and unpinned remote template download makes this high risk for agent deployment.
Confidence: 89%Severity: 81%
Audit Metadata