gm
Warn
Audited by Socket on Sep 20, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS. The skill’s capabilities mostly fit its stated orchestration purpose, but its trust model relies on remote execution of same-org GitHub installer scripts and an external runner binary. That is coherent with the product design and partially mitigated by same-publisher docs and checksum verification, yet still creates meaningful supply-chain risk without evidence of overt credential theft or malicious exfiltration.
Confidence: 86%Severity: 58%
Audit Metadata