ralph
Warn
Audited by Socket on Nov 6, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
The skill’s core development workflow is consistent with its stated purpose, and the excerpt shows no credential harvesting or data exfiltration. Risk comes chiefly from persistent autonomous editing, reduced confirmation prompts in YOLO mode, and an unpinned transitive skill source; review the omitted scripts and use explicit approval and pinned sources.
Confidence: 94%Severity: 56%
Audit Metadata