gemini-cli
Warn
Audited by Socket on Apr 2, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
BENIGN with elevated operational risk. The skill is coherent with its stated purpose and uses an official Google CLI and documented auth paths, but it delegates broad autonomous actions to a secondary agent and exposes a high-risk `--yolo` auto-approval mode. Main concerns are autonomy and unpinned global install, not malware or deceptive data routing.
Confidence: 87%Severity: 68%
Audit Metadata