fdk-uat
Warn
Audited by Socket on Oct 14, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS: the skill’s capabilities broadly fit release acceptance testing, but its trust model is risky because the main action is executing a mutable raw GitHub bootstrap script via pipe-to-shell, then performing high-impact git remote changes. No clear credential harvesting or off-purpose exfiltration is shown, and data flows stay mainly within GitHub/Orca, but the install path and remote-write scope are stronger than ideal for a testing skill.
Confidence: 90%Severity: 62%
Audit Metadata