auto-polish
Warn
Audited by Socket on Sep 25, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS: the skill’s stated UI-polish purpose broadly matches its actions, and there is no strong malware or installer abuse signal. However, it is high-friction and overreaching for a skill because it auto-runs without being asked, scans a broad set of local agent tooling, starts local services/browser flows, and may forward project screenshots to third-party tutor CLIs.
Confidence: 84%Severity: 58%
Audit Metadata