full-send
Warn
Audited by Socket on Nov 3, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
The skill is coherent as an end-to-end PR automation workflow, and the available evidence does not indicate credential theft, exfiltration, or malicious payloads. Its main risk is delegated autonomy to modify, force-push, and merge code without per-action approval, compounded by reliance on other skills and exposure to untrusted PR content.
Confidence: 91%Severity: 68%
Audit Metadata