first-ask
Warn
Audited by Socket on Apr 14, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS. The skill's purpose and behavior are mostly coherent, and it does not request credentials or route data to obvious exfiltration endpoints. Risk comes from third-party tool dependence, remote skill-loading context, and optional web research feeding into later actions; this looks more like a moderately risky orchestration skill than malicious content.
Confidence: 80%Severity: 57%
Audit Metadata