content-ideation
Warn
Audited by Socket on Aug 27, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS. The stated purpose is coherent with the listed research capabilities, and scope is otherwise narrow and read-only. However, the skill routes all platform access through SandBase-managed tools instead of direct official APIs, creating an unnecessary intermediary data path for user queries and any associated account credentials. With no local access, execution, or credential-file harvesting shown, this is not confirmed malware, but the proxy-style data flow makes the skill medium risk.
Confidence: 87%Severity: 56%
Audit Metadata