looker-performance-optimizer
Warn
Audited by Socket on Jul 27, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS. The skill’s core capabilities largely fit Looker performance tuning, but it grants an AI agent high-impact autonomy: executing diagnostics on production-connected systems, editing LookML, and deploying to production. Tooling/data flows are mostly legitimate for the purpose, yet the combination of autonomous production actions, unpinned external tooling references, and processing untrusted SQL/plan content raises the overall risk above benign.
Confidence: 87%Severity: 72%
Audit Metadata