reviewer
Warn
Audited by Socket on Oct 2, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS: The skill's purpose is coherent, and the external CLI appears to be Google's official agy tool, but the trust footprint is broad: unsandboxed execution, stored OAuth access, delegation to an external native CLI, unrestricted default behavior, and acknowledged prompt-injection exposure from untrusted code. This looks like a legitimate but high-trust workflow integration rather than malware.
Confidence: 86%Severity: 69%
Audit Metadata